[NF4][FSDP] return contiguous quantization_factor

[weifengpy]

before, NF4.quantization_factor is non-contiguous, meaning NF4.stride() is (16, 0)
after, NF4.quantization_factor contiguous with NF4.stride() is (1, 0). it's easy to work with FSDP since we do .view(unit8) and it requries contiguous tensor

import torch
from torchao.dtypes.nf4tensor import to_nf4
inpt_tensor = torch.rand(4096, device='cuda')
inpt_tensor_nf4 = to_nf4(inpt_tensor, 32, 16)
print(inpt_tensor_nf4.quantization_factor.stride())

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/298

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Upgrade MacOS runner to 14

✅ No Failures

As of commit ce26526 with merge base 68ce5b8 ():
💚 Looks good so far! There are no failures yet. 💚

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

[drisspg]

🚢