[Data] Decrease scale of map_groups release test

[bveeramani]

Description

#58234 increased the scale of the map_groups release test from SF 10 to SF 100. Since then, the release test has been consistently failing (see #58312).

To avoid a perpetually broken release test, this PR reverts the scale to SF 10 while we investigate and fix the scalability issue.

Related issues

#58312

[gemini-code-assist]

Code Review

This pull request correctly reverts the scale factor for the map_groups release test from SF 100 to SF 10, which is a necessary step to resolve the consistent test failures. My review includes a suggestion to add a TODO comment to track this temporary change, ensuring it's revisited once the underlying scalability issue is fixed. This will improve the long-term maintainability of the test configuration.

[gemini-code-assist]

This change is a good temporary fix to unblock the release tests. To improve long-term maintainability, it would be beneficial to add a TODO comment in the code explaining why the scale factor was reduced and referencing the tracking issue (#58312).

Since this script block uses the folded block scalar (>), adding an inline comment will comment out the rest of the script. You could change the block scalar to literal style (|) on line 195 to add the comment on a separate line.

For example:

  run:
    timeout: 3600
    script: |
      # TODO(#58312): Revert to --sf 100 once the scalability issue is fixed.
      python groupby_benchmark.py --sf 10 --map-groups --group-by {{columns}}
      --shuffle-strategy {{shuffle_strategy}}

This will ensure the reason for this temporary change is not lost over time.