Added a new without_checkpoint_model:bool param to trainer.train()

[stevemadere]

Added a new option without_checkpoint_model:bool to the Trainer.train() params.

This prevents re-loading the model supplied to the constructor during a train(resume_from_checkpoint) operation so that models that the Trainer does not explicitly know how to load itself can still be resumed from a checkpoint.

Fixes #29740

Before submitting

• [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• [x] 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.
• [x] 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?

@muellerzr , @pacman100, @amyeroberts

[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.

[amyeroberts]

hmmm, I'm not sure about this. This seems like a workaround that avoids addressing a problem with the trainer.

~Could you give an example of a model that's saved out with trainer but then can't resume from checkpoint? Ideally as a small script to reproduce~ Never mind - I just saw the details in the issue :)