feat: Change paged FP8 prefill back to regular attention

[ani300]

Description of the change

Deeptools developers prefer the prefill for paged fp8 attention to be regular attention instead of the math fp8 attention, given that the FP8 projections return FP16 anyways and running attention in FP8 for prefill essentially requires casting activations to FP8 twice.

Related issues or PRs

Internal discussion.

How to verify the PR

Was the PR tested

• I have added >=1 unit test(s) for every new method I have added (if that coverage is difficult, please briefly explain the reason)
• I have ensured all unit tests pass

Checklist for passing CI/CD:

• All commits are signed showing "Signed-off-by: Name <email@domain.com>" with git commit -signoff or equivalent
• PR title and commit messages adhere to Conventional Commits
• Contribution is formatted with tox -e fix
• Contribution passes linting with tox -e lint
• Contribution passes spellcheck with tox -e spellcheck
• Contribution passes all unit tests with tox -e unit

Note: CI/CD performs unit tests on multiple versions of Python from a fresh install. There may be differences with your local environment and the test environment.

[andrea-fasoli]

followed the discussion. Change looks good.