fix: fix fp8dqrow setting

[nifleisch]

Description

The quant type "fp8dqrow" was not working. This fix makes this option work for GPUs with CUDA Compute Capability >= 8.9 (e.g., L40S, H100).

Type of Change

• 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)
• This change requires a documentation update

How Has This Been Tested?

Run the algorithms with quant type "fp8dqrow" and "int8dq". Tested the combination with torch.compile and without it.

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• 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

[johnrachwan123]

LGTM, can we add which GPUs are supported in the PR description please ?

[llcnt]

Looks good to me, thx :)