fix(metrics): Remove use-metrics-layer flag from endpoints and snuba signatures (#108752)

## Summary

The `organizations:use-metrics-layer` feature flag is permanently
disabled — it defaults to `False` everywhere and is not enabled for any
organization via flagpole. This PR removes all references from:

- **API endpoints**: `organization_events.py`,
`organization_events_stats.py`, `organization_events_meta.py`,
`organization_events_histogram.py`, `data_export.py`
- **Snuba module function signatures**: All dataset modules
(`discover.py`, `metrics_performance.py`,
`metrics_enhanced_performance.py`, `spans_metrics.py`,
`spans_indexed.py`, `functions.py`, `errors.py`, `transactions.py`,
`profiles.py`, `issue_platform.py`, `entity_subscription.py`,
`types.py`)
- **Other**: `snuba/tasks.py` telemetry tag,
`check_am2_compatibility.py` QueryBuilderConfig kwarg
- **Dead test classes**: 4 `WithMetricLayer` test classes that only
tested the disabled code path

**PR 1 of 3** — follow-up PRs will:
- PR2: Remove from the query builder core (`MetricsQueryBuilder`,
`MetricsLayerDatasetConfig`, etc.)
- PR3: Remove flag registration from `temporary.py` and flagpole config

## Test plan
- [x] `pre-commit run --files` passes on all changed files
- [ ] CI passes
- [ ] Verify no remaining `use_metrics_layer` references in modified
files via grep

Author: wedamija

src/sentry/api/endpoints/organization_events.py

@@ -111,7 +111,6 @@ def get_features(self, organization: Organization, request: Request) -> Mapping[
             "organizations:performance-use-metrics",
             "organizations:profiling",
             "organizations:dynamic-sampling",
-            "organizations:use-metrics-layer",
             "organizations:starfish-view",
             "organizations:on-demand-metrics-extraction",
             "organizations:on-demand-metrics-extraction-widgets",
@@ -274,7 +273,6 @@ def _data_fn(
                 transform_alias_to_input_format=True,
                 # Whether the flag is enabled or not, regardless of the referrer
                 has_metrics=use_metrics,
-                use_metrics_layer=batch_features.get("organizations:use-metrics-layer", False),
                 on_demand_metrics_enabled=on_demand_metrics_enabled,
                 on_demand_metrics_type=on_demand_metrics_type,
                 fallback_to_transactions=True,

src/sentry/api/endpoints/organization_events_histogram.py

@@ -65,9 +65,6 @@ def get(self, request: Request, organization: Organization) -> Response:
         use_metrics = features.has(
             "organizations:performance-use-metrics", organization=organization, actor=request.user
         )
-        use_metrics_layer = features.has(
-            "organizations:use-metrics-layer", organization=organization, actor=request.user
-        )
         dataset = self.get_dataset(request) if use_metrics else discover
         metrics_enhanced = dataset != discover
 
@@ -89,7 +86,6 @@ def get(self, request: Request, organization: Organization) -> Response:
                         max_value=data.get("max"),
                         data_filter=data.get("dataFilter"),
                         referrer="api.organization-events-histogram",
-                        use_metrics_layer=use_metrics_layer,
                     )
 
                 return Response(results)

src/sentry/api/endpoints/organization_events_meta.py

@@ -39,7 +39,6 @@ def get_features(self, organization: Organization, request: Request) -> dict[str
             "organizations:performance-use-metrics",
             "organizations:profiling",
             "organizations:dynamic-sampling",
-            "organizations:use-metrics-layer",
             "organizations:starfish-view",
         ]
         batch_features = features.batch_has(
@@ -102,7 +101,6 @@ def get(self, request: Request, organization: Organization) -> Response:
                     query=request.query_params.get("query"),
                     referrer=Referrer.API_ORGANIZATION_EVENTS_META.value,
                     has_metrics=use_metrics,
-                    use_metrics_layer=batch_features.get("organizations:use-metrics-layer", False),
                     # TODO: @athena - add query_source when all datasets support it
                     # query_source=(
                     #     QuerySource.FRONTEND if is_frontend_request(request) else QuerySource.API

src/sentry/api/endpoints/organization_events_stats.py

@@ -70,7 +70,6 @@ def get_features(
             "organizations:performance-use-metrics",
             "organizations:dashboards-mep",
             "organizations:mep-rollout-flag",
-            "organizations:use-metrics-layer",
             "organizations:starfish-view",
             "organizations:on-demand-metrics-extraction",
             "organizations:on-demand-metrics-extraction-widgets",
@@ -219,7 +218,6 @@ def get(self, request: Request, organization: Organization) -> Response:
             metric_types = ",".join(metric_type_values)
             return Response({"detail": f"Metric type must be one of: {metric_types}"}, status=400)
 
-        force_metrics_layer = request.GET.get("forceMetricsLayer") == "true"
         use_rpc = dataset in RPC_DATASETS
         transform_alias_to_input_format = (
             request.GET.get("transformAliasToInputFormat") == "1" or use_rpc
@@ -343,11 +341,6 @@ def get_rpc_config():
                 comparison_delta=comparison_delta,
                 allow_metric_aggregates=allow_metric_aggregates,
                 has_metrics=use_metrics,
-                # We want to allow people to force use the new metrics layer in the query builder. We decided to go for
-                # this approach so that we can have only a subset of parts of sentry that use the new metrics layer for
-                # their queries since right now the metrics layer has not full feature parity with the query builder.
-                use_metrics_layer=force_metrics_layer
-                or batch_features.get("organizations:use-metrics-layer", False),
                 on_demand_metrics_enabled=use_on_demand_metrics
                 and (
                     batch_features.get("organizations:on-demand-metrics-extraction", False)

src/sentry/data_export/endpoints/data_export.py

@@ -252,7 +252,6 @@ def get_features(self, organization: Organization, request: Request) -> dict[str
             "organizations:performance-use-metrics",
             "organizations:profiling",
             "organizations:dynamic-sampling",
-            "organizations:use-metrics-layer",
             "organizations:starfish-view",
         ]
         batch_features = features.batch_has(

src/sentry/snuba/discover.py

@@ -178,7 +178,6 @@ def query(
     transform_alias_to_input_format: bool = False,
     sample: float | None = None,
     has_metrics: bool = False,
-    use_metrics_layer: bool = False,
     skip_tag_resolution: bool = False,
     extra_columns: list[Column] | None = None,
     on_demand_metrics_enabled: bool = False,
@@ -276,7 +275,6 @@ def timeseries_query(
     functions_acl: list[str] | None = None,
     allow_metric_aggregates: bool = False,
     has_metrics: bool = False,
-    use_metrics_layer: bool = False,
     on_demand_metrics_enabled: bool = False,
     on_demand_metrics_type: MetricSpecType | None = None,
     dataset: Dataset = Dataset.Discover,
@@ -822,7 +820,6 @@ def spans_histogram_query(
     limit_by: list[str] | None = None,
     extra_condition: list[Condition] | None = None,
     normalize_results: bool = True,
-    use_metrics_layer: bool = False,
     on_demand_metrics_enabled: bool = False,
     on_demand_metrics_type: MetricSpecType | None = None,
     query_source: QuerySource | None = None,
@@ -913,7 +910,6 @@ def histogram_query(
     histogram_rows: int | None = None,
     extra_conditions: list[Condition] | None = None,
     normalize_results: bool = True,
-    use_metrics_layer: bool = False,
     on_demand_metrics_enabled: bool = False,
     on_demand_metrics_type: MetricSpecType | None = None,
     query_source: QuerySource | None = None,

src/sentry/snuba/entity_subscription.py

@@ -386,7 +386,6 @@ def __init__(
             )
         self.org_id = extra_fields["org_id"]
         self.time_window = time_window
-        self.use_metrics_layer = False
 
         self.on_demand_metrics_enabled = features.has(
             "organizations:on-demand-metrics-extraction",
@@ -418,15 +417,9 @@ def _get_use_case_id(self) -> UseCaseID:
             return UseCaseID.SESSIONS
 
     def resolve_tag_key_if_needed(self, string: str) -> str:
-        if self.use_metrics_layer:
-            return string
-
         return resolve_tag_key(self._get_use_case_id(), self.org_id, string)
 
     def resolve_tag_values_if_needed(self, strings: Sequence[str]) -> Sequence[str | int]:
-        if self.use_metrics_layer:
-            return strings
-
         return resolve_tag_values(self._get_use_case_id(), self.org_id, strings)
 
     def build_query_builder(
@@ -456,7 +449,6 @@ def build_query_builder(
             time_range_window=self.time_window,
             config=QueryBuilderConfig(
                 skip_time_conditions=True,
-                use_metrics_layer=self.use_metrics_layer,
                 on_demand_metrics_enabled=self.on_demand_metrics_enabled,
                 on_demand_metrics_type=MetricSpecType.SIMPLE_QUERY,
                 skip_field_validation_for_entity_subscription_deletion=skip_field_validation_for_entity_subscription_deletion,
@@ -546,17 +538,13 @@ def aggregate_query_results(
         return aggregated_results
 
     def get_snql_extra_conditions(self) -> list[Condition]:
-        # If we don't use the metrics layer we need to filter by metric here. The metrics layer does this automatically.
-        if not self.use_metrics_layer:
-            return [
-                Condition(
-                    Column("metric_id"),
-                    Op.EQ,
-                    resolve(UseCaseID.SESSIONS, self.org_id, self.metric_key.value),
-                )
-            ]
-
-        return []
+        return [
+            Condition(
+                Column("metric_id"),
+                Op.EQ,
+                resolve(UseCaseID.SESSIONS, self.org_id, self.metric_key.value),
+            )
+        ]
 
 
 class MetricsCountersEntitySubscription(BaseCrashRateMetricsEntitySubscription):

src/sentry/snuba/errors.py

@@ -56,7 +56,6 @@ def query(
     transform_alias_to_input_format: bool = False,
     sample: float | None = None,
     has_metrics: bool = False,
-    use_metrics_layer: bool = False,
     skip_tag_resolution: bool = False,
     extra_columns: list[Column] | None = None,
     on_demand_metrics_enabled: bool = False,
@@ -112,7 +111,6 @@ def timeseries_query(
     functions_acl: list[str] | None = None,
     allow_metric_aggregates=False,
     has_metrics=False,
-    use_metrics_layer=False,
     on_demand_metrics_enabled=False,
     on_demand_metrics_type: MetricSpecType | None = None,
     query_source: QuerySource | None = None,

src/sentry/snuba/functions.py

@@ -37,7 +37,6 @@ def query(
     transform_alias_to_input_format: bool = False,
     has_metrics: bool = False,
     functions_acl: list[str] | None = None,
-    use_metrics_layer: bool = False,
     on_demand_metrics_enabled: bool = False,
     on_demand_metrics_type: MetricSpecType | None = None,
     fallback_to_transactions=False,
@@ -87,7 +86,6 @@ def timeseries_query(
     functions_acl: list[str] | None = None,
     allow_metric_aggregates: bool = False,
     has_metrics: bool = False,
-    use_metrics_layer: bool = False,
     on_demand_metrics_enabled: bool = False,
     on_demand_metrics_type: MetricSpecType | None = None,
     query_source: QuerySource | None = None,

src/sentry/snuba/issue_platform.py

@@ -33,7 +33,6 @@ def query(
     transform_alias_to_input_format=False,
     sample=None,
     has_metrics=False,
-    use_metrics_layer=False,
     skip_tag_resolution=False,
     on_demand_metrics_enabled=False,
     on_demand_metrics_type: MetricSpecType | None = None,
@@ -117,7 +116,6 @@ def timeseries_query(
     functions_acl: list[str] | None = None,
     allow_metric_aggregates=False,
     has_metrics=False,
-    use_metrics_layer=False,
     on_demand_metrics_enabled=False,
     on_demand_metrics_type: MetricSpecType | None = None,
     query_source: QuerySource | None = None,