Qwen3.6 27B-MTP / Vulkan native PP2048 TG128, MTP off

User reports Qwen3.6 at 13.8 t/s generation and 294 t/s prompt processing on an AMD Strix Halo 128GB rig. Setup is the charlie12345/rocmfp4-llama fork with the ROCmFP4 STRIX hybrid embF16/headQ6 quant, run through native llama-bench with -ngl 999 -fa 1 -ub 1024 -r 3 and no MTP. The 13.8 t/s is the Vulkan0 TG128 figure and 294 t/s the Vulkan0 PP2048 figure, separate synthetic tests rather than one request; batch size, KV types, context depth and resident memory are unpublished.