[Chat][Bug][Feature]Add Peft support and fix the bug when ptx_coef is enabled

[yynil]

[Chat][Bug][Feature]Add Peft support and fix the bug when ptx_coef is enabled

📌 Checklist before creating the PR

• [ x] I have created an issue for this PR for traceability
• The title follows the standard format: [doc/gemini/tensor/...]: A concise description
• I have added relevant tags if possible for us to better distinguish different PRs

🚨 Issue number

issues/3432
issues/3454

📝 What does this PR do?

1. Fix bug of wrong data fetching when pretrained data loss is enabled.
2. Add Huggingface Peft model support for SFT and Prompts(aka. Stage-1 and Stage-3) training.

💥 Checklist before requesting a review

• [ x] I have linked my PR to an issue (instruction)
• [ x] My issue clearly describes the problem/feature/proposal, with diagrams/charts/table/code if possible
• [ x] I have performed a self-review of my code
• [ x] I have added thorough tests.
• [ x] I have added docstrings for all the functions/methods I implemented

⭐️ Do you enjoy contributing to Colossal-AI?

• [ x] 🌝 Yes, I do.
• 🌚 No, I don't.

Tell us more if you don't enjoy contributing to Colossal-AI.

[ht-zhou]

Thank you for your nice work .
We suggest you to make a folder under example folder named "community" to store your code there.
Or you can choose to add a flag named "peft".
Also, we suggest you to delete the Chinese comment.

[yynil]

Thank you for your nice work . We suggest you to make a folder under example folder named "community" to store your code there. Or you can choose to add a flag named "peft". Also, we suggest you to delete the Chinese comment.

Very nice suggestions. I just move all my codes to community folder and remove the Chinese comments while keeping the original files untackled.