[PyTorch] Fix L3 FA tests

[cyanguwa]

Description

This PR fixes the L3 tests for FP8 current scaling in L3_pytorch_FA_versions_test. The fix is only related to the selection logic in the test and not the backend support itself.

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 refactoring

Changes

Please list the changes introduced in this PR:

• L3 CI test fix.

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

[greptile-apps]

Greptile Summary

Fixed the test_dpa_fp8_vs_f16 test to properly handle backend availability checks, preventing undefined variable errors when FusedAttention backends aren't supported.

Key changes:

• Separated FP8 and F16 backend availability tracking (fused_attn_supported_fp8 vs fused_attn_supported_f16)
• Wrapped FusedAttention test execution in conditional blocks that check for backend support
• Updated all comparison blocks to verify both result sets exist before accessing them

Impact:
The fix prevents NameError exceptions when running tests on hardware that doesn't support certain attention backends (particularly relevant for FP8 current scaling mode). Tests now gracefully skip unsupported configurations instead of crashing.

Confidence Score: 5/5

• This PR is safe to merge with no risk
• Test-only change that fixes a real bug with proper conditional guards. All variable accesses are correctly guarded by backend availability checks, preventing undefined variable errors. The logic is sound and handles all edge cases properly.
• No files require special attention

Important Files Changed

Filename	Overview
tests/pytorch/attention/test_attention.py	Fixed FP8 attention backend availability checking to prevent undefined variable errors when backends aren't supported

_{Last reviewed commit: 1a40989}

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[cyanguwa]

/te-ci pytorch L3