[Kernel][Hardware][AMD] Bf16 mfma opt for ROCm skinny GEMMs (#17071)

Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
Signed-off-by: charlifu <charlifu@amd.com>
Co-authored-by: charlifu <charlifu@amd.com>

vllm/platforms/rocm.py

@@ -104,6 +104,7 @@ def device_id_to_physical_device_id(device_id: int) -> int:
         return device_id
 
 
+@cache
 def on_mi250_mi300() -> bool:
     GPU_ARCH = torch.cuda.get_device_properties("cuda").gcnArchName
     return any(arch in GPU_ARCH for arch in ["gfx90a", "gfx942"])