Hg model training with torch DataLoader

[annasun28]

What does this PR do? Please describe:

• For the initial SFT recipe, we've set it up with the dataloader / preprocessing logic from LLaMa-factory which relies on torch DataLoader. TODO: handle state dict saving w/stateful sampler. TBD: create a new class instead of overloading the existing DataPipelineReader with 2 different pipeline types
• Update fairseq2 HgTokenizer with some methods required from the base transformers Tokenizer. TBD: whether we should actually subclass and override methods so that we automatically fall back to the base transformers Tokenizer methods?
• Enable FSDP for Hg model (matching the way FSDP wrapping is handled in transformers / accelerate). TODO: make the transformer_cls_names_to_wrap configurable

Tested with https://github.com/fairinternal/seamless_next/pull/750

Does your PR introduce any breaking changes? If yes, please list them: N/A

Check list:

• [ ] Was the content of this PR discussed and approved via a GitHub issue? (no need for typos or documentation improvements)
• [ ] Did you read the contributor guideline?
• [ ] Did you make sure that your PR does only one thing instead of bundling different changes together?
• [ ] Did you make sure to update the documentation with your changes? (if necessary)
• [ ] Did you write any new necessary tests?
• [ ] Did you verify new and existing tests pass locally with your changes?
• [ ] Did you update the CHANGELOG? (no need for typos, documentation, or minor internal changes)