[mxfp8] fix test nan != nan issue

[danielvegamyhre]

• Test was failing due to the counter-intuitive "nan != nan"
• 255 is e8m0 representation of nan, so the ground_truth_scale is nan.
• The mxfp8 conversion produces a nan scale as well, but the assert_close fails because nan != nan
• This PR changes the assertion to just validate the scale is nan as expected

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3273

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 1 Pending

As of commit 082a7e4 with merge base 1e473ed ():

NEW FAILURE - The following job has failed:

• Run 1xH100 Tests / test (H100, linux.aws.h100, --pre torch torchvision torchaudio fbgemm-gpu-genai --index-url https... / linux-job (gh)
  test_expected_gpu_kernel_fbgemm

This comment was automatically generated by Dr. CI and updates every 15 minutes.