ESQL: Use DEFAULT_UNSORTABLE topN encoder for the TSID_DATA_TYPE

Author: leontyevdv

x-pack/plugin/esql/qa/testFixtures/src/main/resources/topN.csv-spec

@@ -168,3 +168,20 @@ FROM employees
 avg_worked_seconds:long | birth_date:date      | emp_no:i | gender:k | height:d | height.float:d     | height.half_float:d | height.scaled_float:d | hire_date:date       | is_rehired:bool      | job_positions:k     | languages:i | languages.byte:i | languages.long:l | languages.short:short | salary:i | salary_change:d | salary_change.int:i | salary_change.keyword:k | salary_change.long:l | still_hired:bool | name:k  | first_name:k | last_name:k
 349086555               | 1961-09-01T00:00:00Z | 10056    | F        | 1.57     | 1.5700000524520874 | 1.5703125           | 1.57                  | 1990-02-01T00:00:00Z | [false, false, true] | [Senior Team Lead]  | 2           | 2                | 2                | 2                     | 33370    | [-5.17, 10.99]  | [-5, 10]            | [-5.17, 10.99]          | [-5, 10]             | true             | Brendon | Bernini      | Brendon
 ;
+
+sortingByTsidMetadataField
+required_capability: metadata_tsid_field
+
+FROM k8s METADATA _tsid
+| SORT events_received ASC
+| LIMIT 5
+| KEEP events_received, @timestamp, cluster, pod
+;
+
+events_received:long | @timestamp:datetime      | cluster:keyword | pod:keyword
+0                    | 2024-05-10T00:04:49.000Z | qa              | one
+0                    | 2024-05-10T00:02:43.000Z | staging         | two
+0                    | 2024-05-10T00:03:13.000Z | staging         | one
+0                    | 2024-05-10T00:07:33.000Z | qa              | two
+0                    | 2024-05-10T00:05:16.000Z | prod            | one
+;

x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/planner/LocalExecutionPlanner.java

@@ -485,9 +485,9 @@ private PhysicalOperation planTopN(TopNExec topNExec, LocalExecutionPlannerConte
                 case VERSION -> TopNEncoder.VERSION;
                 case DOC_DATA_TYPE -> new DocVectorEncoder(context.shardContexts);
                 case BOOLEAN, NULL, BYTE, SHORT, INTEGER, LONG, DOUBLE, FLOAT, HALF_FLOAT, DATETIME, DATE_NANOS, DATE_PERIOD, TIME_DURATION,
-                    OBJECT, SCALED_FLOAT, UNSIGNED_LONG, TSID_DATA_TYPE -> TopNEncoder.DEFAULT_SORTABLE;
+                    OBJECT, SCALED_FLOAT, UNSIGNED_LONG -> TopNEncoder.DEFAULT_SORTABLE;
                 case GEO_POINT, CARTESIAN_POINT, GEO_SHAPE, CARTESIAN_SHAPE, COUNTER_LONG, COUNTER_INTEGER, COUNTER_DOUBLE, SOURCE,
-                    AGGREGATE_METRIC_DOUBLE, DENSE_VECTOR, GEOHASH, GEOTILE, GEOHEX, EXPONENTIAL_HISTOGRAM ->
+                    AGGREGATE_METRIC_DOUBLE, DENSE_VECTOR, GEOHASH, GEOTILE, GEOHEX, EXPONENTIAL_HISTOGRAM, TSID_DATA_TYPE ->
                     TopNEncoder.DEFAULT_UNSORTABLE;
                 // unsupported fields are encoded as BytesRef, we'll use the same encoder; all values should be null at this point
                 case PARTIAL_AGG, UNSUPPORTED -> TopNEncoder.UNSUPPORTED;