[GLUTEN-11599][VL] Support approx_count_distinct_for_intervals function

[malinjawi]

Summary

• Map Spark approx_count_distinct_for_intervals to the Velox backend aggregate

Context

• Fixes [VL] Add support for approx_count_distinct_for_intervals #11599
• Depends on feat(spark): Add approx_count_distinct_for_intervals aggregate function facebookincubator/velox#16595

Testing

• Not run (mapping‑only change)

[github-actions]

Run Gluten Clickhouse CI on x86

[github-actions]

Run Gluten Clickhouse CI on x86

[jinchengchenghh]

Please update here to verify you Velox PR, https://github.com/apache/incubator-gluten/blob/main/ep/build-velox/src/get-velox.sh#L28, and make sure your native implementation is really called. Provide an example test in CI, e.g. warning message disappeared.

[main] WARN org.apache.spark.sql.execution.GlutenFallbackReporter - Validation failed for plan: ObjectHashAggregate[QueryId=113], due to: 
 - Validation failed with exception from: RegularHashAggregateExecTransformer, reason: Could not find a valid substrait mapping name for approx_count_distinct_for_intervals(s_store_sk#87088L, org.apache.spark.sql.catalyst.expressions.UnsafeArrayData@3915186b, 0.05, 0, 0).

[github-actions]

Run Gluten Clickhouse CI on x86

[malinjawi]

Please update here to verify you Velox PR, https://github.com/apache/incubator-gluten/blob/main/ep/build-velox/src/get-velox.sh#L28, and make sure your native implementation is really called. Provide an example test in CI, e.g. warning message disappeared.

[main] WARN org.apache.spark.sql.execution.GlutenFallbackReporter - Validation failed for plan: ObjectHashAggregate[QueryId=113], due to: 
 - Validation failed with exception from: RegularHashAggregateExecTransformer, reason: Could not find a valid substrait mapping name for approx_count_distinct_for_intervals(s_store_sk#87088L, org.apache.spark.sql.catalyst.expressions.UnsafeArrayData@3915186b, 0.05, 0, 0).

Updated locally to address this review. I pinned Gluten’s Velox fetch script to verify against Velox PR 16595 in get-velox.sh, and I added a Spark 3.5 fallback regression that runs ANALYZE TABLE ... COMPUTE STATISTICS FOR COLUMNS and asserts there are 0 fallback nodes. This covers the previous approx_count_distinct_for_intervals validation failure path ("Could not find a valid substrait mapping name ...") and verifies the native implementation is used instead of falling back.

[github-actions]

Run Gluten Clickhouse CI on x86

[github-actions]

Run Gluten Clickhouse CI on x86

[github-actions]

Run Gluten Clickhouse CI on x86

[github-actions]

This PR is stale because it has been open 45 days with no activity. Remove stale label or comment or this will be closed in 10 days.

[github-actions]

This PR was auto-closed because it has been stalled for 10 days with no activity. Please feel free to reopen if it is still valid. Thanks.