This PR refactors the monolithic gke-a3-ultragpu integration test into two distinct, focused tests: a lightweight Kueue validation test and a heavyweight NCCL performance test.
This change is critical to address a series of severe stability issues discovered during the Kueue-on-Helm migration. The previous, single-test structure was found to be brittle and was the root cause of CI failures where the cluster would self-destruct by scaling its own GPU nodes to zero. This new structure is more stable, provides faster feedback, is more resource-efficient, and completely avoids this failure.

Submission Checklist

NOTE: Community submissions can take up to 2 weeks to be reviewed.

Please take the following actions before submitting this pull request.

• Fork your PR branch from the Toolkit "develop" branch (not main)
• Test all changes with pre-commit in a local branch #
• Confirm that "make tests" passes all tests
• Add or modify unit tests to cover code changes
• Ensure that unit test coverage remains above 80%
• Update all applicable documentation
• Follow Cluster Toolkit Contribution guidelines #