feat: Merge kernels from vLLM and FlashInfer

[drunkcoding]

Description

Fuse MoE layer kernels

Motivation

Kernel launch overhead too large

Type of Change

• [ ] Bug fix
• [x] New feature
• [x] Breaking change
• [x] Documentation update

Checklist

• [x] I have read the CONTRIBUTION guide.
• [x] I have updated the tests (if applicable).
• [x] I have updated the documentation (if applicable).