[LLM] Add quantize_kv optimization for yuan2 model

[sgwhat]

Description

Add quantize_kv to optimize the performance of yuan2 model on mtl.

1. Why the change?

2. User API changes

3. Summary of the change

4. How to test?

• [ ] N/A
• [ ] Unit test
• [ ] Application test
• [ ] Document test
• [ ] ...

[sgwhat]

The Yuan2 model requires quantize_kv support for the head_dim of 64, which only supports 128 for now. I am currently working on it.

[qiuxin2012]

Fix style please.