Use minimal CUDA container for PyTorch GitHub build

[timmoon10]

Description

We've experienced test failures in our GitHub PyTorch build tests since August 2 (success at 10:59:49 GMT, failure at 15:37:12 GMT). OOM errors happen while loading the 24.05 NGC PyTorch container (used since June with #919). It doesn't seem like anything has changed in the container (the downloaded hashes are the same), so I suspect something has changed in the GitHub runners. It's not the first time we've run into memory issues with the PyTorch container (see #462).

This PR modifies the PyTorch build test so it uses a minimal CUDA container, similar to the core build test.

Type of change

• Documentation change (change only to the documentation, either a fix or a new content)
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Infra/Build change
• Code refractor
• Testing

Changes

• Use CUDA container for PyTorch GitHub build

Checklist:

• I have read and followed the contributing guidelines
• The functionality is complete
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes