Add checks for HLG layers in dask-cudf groupby tests

[charlesbluca]

This PR adds helper function check_groupby_result to dask-cudf's groupby tests, and is used in the basic tests to ensure that we are using dask-cudf's groupby_agg function to compute the result as expected.

I also expanded test_groupby_agg to test all supported aggregations, and removed tests that were made superfluous by this change.

[codecov]

Codecov Report

Base: 88.02% // Head: 88.05% // Increases project coverage by +0.02% 🎉

Coverage data is based on head (c1045bd) compared to base (11b875b).
Patch has no changes to coverable lines.

❗ Current head c1045bd differs from pull request most recent head 78be977. Consider uploading reports for the commit 78be977 to get more accurate results

Additional details and impacted files

@@               Coverage Diff                @@
##           branch-22.12   #10853      +/-   ##
================================================
+ Coverage         88.02%   88.05%   +0.02%     
================================================
  Files               135      135              
  Lines             22025    22025              
================================================
+ Hits              19388    19394       +6     
+ Misses             2637     2631       -6     

Impacted Files	Coverage Δ
python/cudf/cudf/core/dataframe.py	93.67% <0.00%> (+0.04%)	⬆️
python/cudf/cudf/core/column/string.py	88.65% <0.00%> (+0.12%)	⬆️
python/cudf/cudf/core/groupby/groupby.py	91.51% <0.00%> (+0.20%)	⬆️
python/cudf/cudf/core/tools/datetimes.py	84.49% <0.00%> (+0.30%)	⬆️
python/cudf/cudf/core/column/lists.py	93.75% <0.00%> (+0.96%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[madsbk]

Looks good, nice clean up @charlesbluca !

[github-actions]

This PR has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this PR if it is no longer required. Otherwise, please respond with a comment indicating any updates. This PR will be labeled inactive-90d if there is no activity in the next 60 days.

[vyasr]

@charlesbluca it looks like this PR is just test improvements and was already approved. Any reason not to merge in the latest 22.12, make sure tests pass, then merge?

[charlesbluca]

It was approved by @madsbk, who IIUC isn't a part of cudf-dask-codeowners; @rjzamora, if you get a chance, would you mind reviewing this (related to discussion we had around improving dask-cudf's groupby testing)?

[wence-]

Minor quibbles. I have a slight concern that the second deleted test reduces coverage of corner-cases, can you explain why not?

[wence-]

@charlesbluca let me know when you think this is ready for another look (e.g. the discussion above is resolved one way or another).

[charlesbluca]

@wence- this should be good for another round of review. The conclusion of the above discussion was to:

• leave the cumulative testing as is, but add an xfail for the failing null cases (opened [BUG] Cumulative aggregation functions fail on dataframe/series groupbys with nulls #12055 to track this)
• remove cumulative aggs from SUPPORTED_AGGS so that we don't use dask-cudf's codepath for them since it doesn't give the correct result; that being said, upstream Dask doesn't support passing cumulative aggs into .agg(), so will want to open an issue there to track adding support for this

Not sure if you think it would be worth it to add in tests for cumulative aggregations using .agg() now, knowing they would be failing, and linking to a related Dask issue?

[wence-]

Two very minor things, but otherwise thanks, LGTM!

[charlesbluca]

@gpucibot merge