CI: add a16w16 GEMM to op tuning workflow

[gyohuangxin]

Summary

• Add the gemm_a16w16 job to the operator tuning script so bf16 GEMM tuning can refresh bf16_tuned_gemm.csv from bf16_untuned_gemm.csv.
• Include gemm_a16w16 in the manual Operators Tuning workflow shape examples.

Test plan

• bash -n .github/scripts/op_tune.sh
• git diff --check

[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 3605 --add-label <label>

[Copilot]

Pull request overview

This PR extends the manual “Operators Tuning” workflow to include the BF16 GEMM tuner (gemm_a16w16) so the workflow can refresh aiter/configs/bf16_tuned_gemm.csv from aiter/configs/bf16_untuned_gemm.csv via the existing op_tune.sh orchestration.

Changes:

• Add a new gemm_a16w16 entry to .github/scripts/op_tune.sh so it can be selected/run like other tuning shapes.
• Update the workflow-dispatch shapes input description to include gemm_a16w16 in the examples.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
.github/workflows/operators-tuning.yaml	Updates the manual workflow input help text to include gemm_a16w16 among example shapes.
.github/scripts/op_tune.sh	Adds gemm_a16w16 to the tuning job list, wiring it to the existing BF16 GEMM untuned/tuned CSV paths.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[yzhou103]

use this, python3 csrc/gemm_a16w16/gemm_tuner.py -i aiter/configs/bf16_untuned_gemm.csv -o aiter/configs/bf16_tuned_gemm.csv. Both are supported, but this can support tune with hipblaslt