[Feature]: support hadamard transform when calibration & tuning

[lkk12014402]

Feature Description

support hadamard transform

Motivation and Use Case

imporve accuracy when do mxfp4 quantization

Alternatives Considered

No response

Definition of Done

No response

Additional Context

No response

[lkk12014402]

push one pull request #1349

[chensuyue]

New PR: #1515

[lkk12014402]

Status:

1. support deterministic_hadamard_matrix and deterministic_hadamard_matrix
2. support inference with transformers with a triton mxfp4 kernel
3. support rtn and autoroud tuning (iters>0)

Todo:

1. align the implementation of SpinQuant: LLM Quantization with Learned Rotations, especially R2/R3, and implement structured fused
2. support online transform for vllm inference
3. support nvfp4
4. verify multi-cards