minor: add sgl-kernel dir

[zhyncs]

Motivation

Currently, we are just creating a new directory first, and will gradually update PyPI, CIs, etc.
It will support torch 2.4, Python 3.9-3.12, CUDA 11.8, 12.1, 12.4. (1*4*3=12)

custom all reduce can move here cc @yizhang2077

Modifications

Checklist

• Format your code according to the Contributor Guide.
• Add unit tests as outlined in the Contributor Guide.
• Update documentation as needed, including docstrings or example tutorials.