Fix hf_device_map device index comparison in prepare_model

[rezaqorbani]

What does this PR do?

This PR fixes an issue where a TypeError is raised if the base model is loaded without specifying device index (e.g. device_map="auto" or device_map="cuda" instead of device_map="cuda:0") for models loaded in 8-bit/4-bit precision with an hf_device_map. The exception is raised in prepare_model() becausetorch.device(device_index) is called with device_index=None.

The fix constructs the device using torch.device(device_type, device_index) instead, which works even when device_index=None.

Added a test to cover the 8-bit/4-bit + hf_device_map path when the model is loaded on CPU (hf_device_map={"": "cpu"}), but the test should also work for other device types as well.

Fixes #3758 and #2357

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@SunMarc @jiqing-feng

[jiqing-feng]

LGTM.

[SunMarc]

Thanks a lot, just a nit

[SunMarc]

Thanks !

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.