fix: remove Designer tags from benchmark datasets (#4899)

Remove azureml.Designer tags from benchmark datasets to prevent them
from appearing in Azure ML Designer UI. These are benchmark-only
datasets not intended for general Designer use.

Updated datasets:
- gpqa_diamond: v1 → v2
- musr: v1 → v2
- truthfulqa: v2 → v3
- frontierscience: v1 → v2
- bigbenchhard: v1 → v2

Updated benchmarkspecs to reference new dataset versions.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: imatiach-msft

assets/benchmarkspecs/builtin/bigbenchhard/spec.yaml

@@ -39,7 +39,7 @@ dataset:
     dataset_source: "https://github.com/suzgunmirac/BIG-Bench-Hard"
   source:
     provider: "mlregistry"
-    sourceDatasetId: "azureml://registries/azureml/data/bigbenchhard/versions/1"
+    sourceDatasetId: "azureml://registries/azureml/data/bigbenchhard/versions/2"
     sourceFormat: "jsonl"
     properties:
       file_name: "bbh_samples_flat.jsonl"

assets/benchmarkspecs/builtin/frontierscience/spec.yaml

@@ -38,7 +38,7 @@ dataset:
     dataset_source: "https://huggingface.co/datasets/openai/frontierscience"
   source:
     provider: "mlregistry"
-    sourceDatasetId: "azureml://registries/azureml/data/frontierscience/versions/1"
+    sourceDatasetId: "azureml://registries/azureml/data/frontierscience/versions/2"
     sourceFormat: "csv"
     properties:
       file_name: "frontierscience.csv"

assets/benchmarkspecs/builtin/gpqa_diamond/spec.yaml

@@ -38,7 +38,7 @@ dataset:
     dataset_source: "https://github.com/idavidrein/gpqa"
   source:
     provider: "mlregistry"
-    sourceDatasetId: "azureml://registries/azureml/data/gpqa_diamond/versions/1"
+    sourceDatasetId: "azureml://registries/azureml/data/gpqa_diamond/versions/2"
     sourceFormat: "csv"
     properties:
       file_name: "gpqa_diamond.csv"

assets/benchmarkspecs/builtin/musr/spec.yaml

@@ -38,7 +38,7 @@ dataset:
     dataset_source: "https://huggingface.co/datasets/TAUR-Lab/MuSR"
   source:
     provider: "mlregistry"
-    sourceDatasetId: "azureml://registries/azureml/data/musr/versions/1"
+    sourceDatasetId: "azureml://registries/azureml/data/musr/versions/2"
     sourceFormat: "csv"
     properties:
       file_name: "MuSR_all.csv"

assets/benchmarkspecs/builtin/truthful_qa/spec.yaml

@@ -33,7 +33,7 @@ dataset:
     dataset_source: "https://github.com/sylinrl/TruthfulQA/"
   source:
     provider: "mlregistry"
-    sourceDatasetId: "azureml://registries/azureml/data/truthful_qa/versions/2"
+    sourceDatasetId: "azureml://registries/azureml/data/truthful_qa/versions/3"
     sourceFormat: "csv"
     properties:
       file_name: "TruthfulQA.csv"

assets/pipelines/data/global_dataset/generic_csv/bigbenchhard/asset.yaml

@@ -1,5 +1,5 @@
 name: bigbenchhard
-version: 1
+version: 2
 type: data
 spec: spec.yaml
 extra_config: storage.yaml

assets/pipelines/data/global_dataset/generic_csv/bigbenchhard/spec.yaml

@@ -2,9 +2,7 @@ $schema: https://azuremlschemas.azureedge.net/latest/data.schema.json
 name: "{{asset.name}}"
 version: "{{asset.version}}"
 tags:
-  azureml.Designer: true
-  azureml.Designer.name: BIG-Bench Hard
-  azureml.Designer.dataTypeId: DataFrameDirectory
+  azureml.Designer: false
 description: BIG-Bench Hard (BBH) benchmark dataset
 type: uri_folder
 path: sample_data/

assets/pipelines/data/global_dataset/generic_csv/frontierscience/asset.yaml

@@ -1,5 +1,5 @@
 name: frontierscience
-version: 1
+version: 2
 type: data
 spec: spec.yaml
 extra_config: storage.yaml

assets/pipelines/data/global_dataset/generic_csv/frontierscience/spec.yaml

@@ -2,9 +2,7 @@ $schema: https://azuremlschemas.azureedge.net/latest/data.schema.json
 name: "{{asset.name}}"
 version: "{{asset.version}}"
 tags:
-  azureml.Designer: true
-  azureml.Designer.name: FrontierScience
-  azureml.Designer.dataTypeId: DataFrameDirectory
+  azureml.Designer: false
 description: FrontierScience benchmark dataset
 type: uri_folder
 path: sample_data/

assets/pipelines/data/global_dataset/generic_csv/gpqa_diamond/asset.yaml

@@ -1,5 +1,5 @@
 name: gpqa_diamond
-version: 1
+version: 2
 type: data
 spec: spec.yaml
 extra_config: storage.yaml