[torchao] Add regex support for ModuleFqnToConfig

[jerryzh168]

Summary:
Similar to pytorch/ao#3084 we added regex support in transformers so people can use regex to quantize the models.

See pytorch/ao#3084 for docs and precedence of different configurations

Uploaded model: https://huggingface.co/torchao-testing/opt-125m-ModuleFqnToConfig-v1-regex-0.14.0.dev

Test Plan:
pytest tests/quantization/torchao_integration/test_torchao.py -k test_module_fqn_to_config_regex

Reviewers:

Subscribers:

Tasks:

Tags:

[Rocketknight1]

cc @MekkCyber for quantization

[jerryzh168]

also cc @SunMarc please take a look

[MekkCyber]

Amazing feature ! Would be nice to add it to the docs for visibility

[SunMarc]

Nice ! Please add some docs around that also. As we are moving to v5 in the next release, maybe we can bump torchao version to clean a but the code @jerryzh168 ?

[jerryzh168]

Nice ! Please add some docs around that also. As we are moving to v5 in the next release, maybe we can bump torchao version to clean a but the code @jerryzh168 ?

yeah we can bump the torchao version, we are preparing for a 0.14 release, we can do it after that I think

[SunMarc]

@bot /style

[github-actions]

Style bot fixed some files and pushed the changes.

[jerryzh168]

@bot /style

[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.

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: torchao_integration