Add LLaVA series to evaluation

[XinrunXu]

No description provided.

[ltzheng]

Have you tested the correctness of the code and the results of Llava models?

[XinrunXu]

I tested on llava-v1.5-7b,llava-v1.5-13b,llava-v1.6-vicuna-7b,llava-v1.6-vicuna-13b, and llava-v1.6-34b.
The code works, but the result is quite low because Lava tends to respond with a 4 tuple instead of a 2 tuple.
And llava-v1.6-mistral-7b cannot work in my code.

[XinrunXu]

haotian-liu/LLaVA#1448
With this issue, my code can work on llava-v1.6-mistral-7b.

[ltzheng]

The code looks good to me but it fails to pass the formatting check. Can you run the format.sh to follow the Coding Style Guide in CONTRIBUTING.md? Thanks!