Support CUDA 12.2

[jameslamb]

Description

• switches to CUDA 12.2.2 for building conda packages and wheels
• adds new tests running against CUDA 12.2.2

Notes for Reviewers

This is part of ongoing work to build and test packages against CUDA 12.2.2 across all of RAPIDS.

For more details see:

• Add support for CUDA 12.2 wheels build-planning#7
• add CUDA 12.2 conda and wheel jobs shared-workflows#166
• adds some dependencies.yaml simplifications missed in refactor CUDA versions in dependencies.yaml #14733

Planning a second round of PRs to revert these references back to a proper branch-24.{nn} release branch of shared-workflows once rapidsai/shared-workflows#166 is merged.

(created with rapids-reviser)

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

[jameslamb]

CUDA 12.2 builds and tests appear to be passing!

The test failures here look unrelated to this change.

ERROR benchmarks/API/bench_functions.py - ValueError: Cannot use `cases=AUTO` in file "bench_functions". `cases=AUTO` is only allowed in files whose name starts with "test_"
============ 468 passed, 9 warnings, 16 errors in 219.08s (0:03:39) ============

(build link)

I see similar failures on other recent pr builds at https://github.com/rapidsai/cudf/actions/workflows/pr.yaml.

[bdice]

#14756 will unblock the failing tests.

[jameslamb]

thanks @bdice !

[jameslamb]

A few tests are failing like this:

FAILED tests/test_groupby.py::test_groupby_basic[non_null-False-count] - TypeError: aggregate() got an unexpected keyword argument 'shuffle_method'
FAILED tests/test_groupby.py::test_groupby_basic[non_null-False-mean] - TypeError: aggregate() got an unexpected keyword argument 'shuffle_method'
FAILED tests/test_groupby.py::test_groupby_basic[non_null-False-std] - TypeError: aggregate() got an unexpected keyword argument 'shuffle_method'
...

(build link)

I suspect that'll be fixed by @rjzamora 's changes in #14762

[jakirkham]

Thanks Bradley! 🙏

Had a couple questions below

[vyasr]

Changes look good and I verified that the build and test jobs are running with the right CUDA packages (channel and version).

[jakirkham]

Toggling to start CI

[bdice]

/merge

[jakirkham]

Rerunning failed test. There was a network error during pip install (times 2)