test: reduce C++ test runtime

[vyasr]

Description

Reduces C++ test latency through two complementary changes:

• Enables the persistent libcudf RTCX/NVRTC cache for C++ test jobs, including workspace-relative cache paths so the cache is restored correctly in GitHub Actions.
• Reduces the runtime of the longest libcudf C++ test fixtures while retaining the behavior each test targets:
  • batches AST cast-expression compilation;
  • reduces oversized Parquet and ORC chunked-reader fixtures while explicitly preserving multi-chunk coverage;
  • scales BGZIP, JSON multi-batch, and compression round-trip inputs to their required boundaries.

The test-fixture changes are organized as atomic commits to make individual reductions easy to evaluate or revert.

Validation

• Ran each affected Google Test binary after its corresponding change.
• Ran the full libcudf CTest suite with the CI-equivalent ctest -j20: 120/120 tests passed in 44.5 seconds locally.

Checklist

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

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[vyasr]

/ok to test

[copy-pr-bot]

/ok to test

@vyasr, there was an error processing your request: E1

See the following link for more information: https://docs.gha-runners.nvidia.com/cpr/e/1/

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[vyasr]

/ok to test

[vyasr]

/ok to test

[vyasr]

/ok to test