[CI/Build][AMD] Skip test on test_hybrid_attention_mamba_tensor_shapes on ROCm, requires FLASHINFER (#29995)

Signed-off-by: Randall Smith <ransmith@amd.com>
Co-authored-by: Randall Smith <ransmith@amd.com>

tests/v1/worker/test_gpu_model_runner.py

@@ -761,6 +761,10 @@ def test_init_kv_cache_with_kv_sharing_valid():
     assert kv_cache_config_after_init.kv_cache_groups[0].layer_names[1] == layer_1
 
 
+@pytest.mark.skipif(
+    current_platform.is_rocm(),
+    reason="Attention backend FLASHINFER is not supported on ROCm.",
+)
 def test_hybrid_attention_mamba_tensor_shapes(monkeypatch):
     """
     The GPU model runner creates different views into the