[Feature] Tune fp8 Gemm and fused moe kernel on B200

[Fridge003]

Checklist

• 1. If the issue you raised is not a feature but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
• 2. Please use English, otherwise it will be closed.

Motivation

The performance of w8a8 gemm kernel and fused moe kernel is not good enough on B200. There is some space for tuning.

Related resources

Reproduction on 8*B200:

python3 -m sglang.bench_one_batch --model-path /dev/shm/DeepSeek-V3 --tp 8 --batch 16 --input-len 1024 --output-len 128 --attention-backend triton --profile

No response

[Fridge003]

fp8 kernels have been tuned in #3272, so no need to tune again.