[gfx1250][FlyDSL] Update a8w8 gemm tuned config

[aoli26]

Motivation

Technical Details

Test Plan

Test Result

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:gfx1250-ffm-triton	Run the five-shard gfx1250 FFM Triton test suite
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 4760 --add-label <label>

[Copilot]

Pull request overview

This PR updates the tuned FlyDSL GEMM configuration entries for gfx1250 in the DSv4 a8w8 blockscale + bpreshuffle config table, adjusting the selected kernel variants and recorded performance metrics for a set of specific GEMM shape rows.

Changes:

• Replaced several gfx1250 tuned rows (notably for larger problem sizes like 512–8192 ranges across multiple shape groups) with new kernel selections (e.g., t128x128x256, t256x128x256, t128x512x128, and updated t256x256x128/256) and updated timing/throughput fields.
• Kept the CSV structure consistent (single entry per key shape examined; no duplicate keys detected for the modified shapes).

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.