Motivation

To fix #1301 and the min p sampling with CUDA graph.

Modifications

• When min p sampling is required, do not use CUDA graph.
• Abolish torch sampler but use flashinfer's sampler, wrapping it into a torch custom operator to make it compatible with torch.compile.

Performance

• Before: 155.51813300924485 tokens/s
• This PR: 156.89880326398756 tokens/s

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.