[SFT] support for ring_attn in SFTTrainer

[kashif]

What does this PR do?

This adds a way to register the ring-attn module instead of the flash-attn for SFT Training

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? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes?
• 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.

[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.

[kashif]

running via on a 2GPU node:

torchrun --nproc_per_node 2 trl/scripts/sft.py --model_name_or_path Qwen/Qwen2-0.5B --dataset_name trl-lib/Capybara --learning_rate 2.0e-5 --num_train_epochs 1 --per_device_train_batch_size 1 --gradient_accumulation_steps 8 --gradient_checkpointing --logging_steps 25 --eval_strategy steps --eval_steps 100 --output_dir Qwen2-0.5B-SFT --sequence_parallel_degree 2 --heads_k_stride 2 --attn-implementation flash_attention_2 --bf16 --torch_dtype bfloat16