[block_wise w8a8]Add block_wise w8w8 tunning scripts

[coolhok]

Motivation

add block-wise int8 w8a8 quantization auto tuning script.

Modifications

1. modify tuning_fused_moe_triton.py adapter block-wise int8 w8a8 moe.
2. add tuning_block_wise_fp8.py for tuning block-wise int8 w8a8 gemm.

run block-wise int8 w8a8 gemm tuning

python tuning_block_wise_w8a8.py -tp 8

other

When running on multiple GPUs, I think we need to split weight_shapes, so each GPU can tune all batch sizes and save complete config.
If splitting batch sizes, each runs the same shape with different batch sizes, but save_config does not support incremental saving.

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://slack.sglang.ai to discuss your PR.

[xihuai18]

Can you upload the tuned config?

[lambert0312]

Can you upload the tuned config?

I can upload a800 tuned config