Add support for the gemma2 language model in multimodal fine-tuning.

[Kidand]

Add support for the gemma2 language model during the process of multimodal model fusion. The fine-tuning process has been successfully completed under 8*H100, hoping to make a contribution.

[czczup]

Hello, thank you for contributing to our repository. We will proceed with the merge after conducting some tests. Thanks again!

[SuHuaFeng]

你好Kidand，我想请教一下更改语言模型的过程，我想把语言部分换成其他模型，我看了您写的代码，不知道该如何去修改，可以分享一下该如何做吗？谢谢！ Hi Kidand, I would like to ask about the process of changing the language model, I want to change the language part to another model, I read the code you wrote, I don't know how to change, can you share how to do it? Thanks!