[PyTorch] Remove MXFP8 scale-inv padding in MXFP8 all-gather

[timmoon10]

Description

#1431 added padding to MXFP8 scale-inverses to satisfy alignment requirements in MXFP8 kernels. However, we didn't properly handle the padding in tensor-parallel all-gathers. This PR makes sure to remove padding when necessary.

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

• Remove MXFP8 scale-inv padding in MXFP8 all-gather

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