Gluon MXFP4 Fuse Reduce Quant

[amd-jrosas]

Motivation

Gluon version of fuse_reduce_rms_mxfp4_quant() to support gfx1250.

Technical Details

Similar layout and design as triton kernel but leverages tdm functions from gfx1250.

Test Plan

Verify passing all test conditions in existing op_test flow for triton kernel. Test case listed below

Parameter	Values	Note
M	1, 32, 256, 8192
N1, N2, N3	(256, 256, 256) and (1536, 512, 64)
SPK	1, 4, 14
shuffle	True, False
padding	True, False
args	"auto", "gluon", "triton"	Default 'auto' unless specified

Test Result

Passes all test cases when explicitly calling gluon and defaults to triton kernel for all other cases.

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[github-actions]

🏷️ CI Guide

Runs automatically on every PR:

• ✅ Pre-checks (submodule verification, code formatting)
• ✅ Aiter op tests (gfx942 + gfx950)
• ✅ Triton tests on MI35X (only when aiter/ops/triton/** or related paths are changed)

Extended tests (opt-in via labels):

Label	Tests
ci:triton-300x	Run an additional Triton test job on MI300X in PRs; main branch always runs both MI35X and MI300X
ci:sglang	SGLang integration tests: DeepSeek-R1-MXFP4 accuracy, Qwen 3.5 accuracy
ci:atom	ATOM benchmark: DeepSeek-R1-0528, GPT-OSS-120B
ci:atom_full	ATOM accuracy suite for PR and main models from ATOM models_accuracy.json
ci:vllm	vLLM benchmark: GPT-OSS-120B, DeepSeek-R1-0528, Kimi-K2.5
ci:all	All standard extended tests (excludes ci:atom_full)

Only add ci:atom_full for FlyDSL or Triton upgrades.
Add labels via the sidebar or gh pr edit 3937 --add-label <label>