Do not initialize torch.distributed process group if one is already initailized

[Yard1]

What does this PR do?

If a torch.distributed process group is already initialized, TrainingArguments will not initialize one by itself. This allows for the user to create their own custom process groups if needed.

Fixes # (issue)

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.

[Yard1]

cc @sgugger

[sgugger]

Thanks a lot for tackling this! LGTM with a tiny nit on the warning message.

[HuggingFaceDocBuilderDev]

The documentation is not available anymore as the PR was closed or merged.