test(serve): skip mm_agg_qwen2-audio-7b vLLM deployment on post-merge (#8411)

Signed-off-by: Dmitry Tokarev <dtokarev@nvidia.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/serve/multimodal_profiles/vllm.py

@@ -85,6 +85,10 @@ VLLM_MULTIMODAL_PROFILES: list[MultimodalModelProfile] = [
         topologies={
             "agg": TopologyConfig(
                 marks=[
+                    pytest.mark.skip(
+                        reason="vLLM engine core init fails on amd64 post-merge. "
+                        "OPS-4445"
+                    ),
                     pytest.mark.post_merge,
                 ],
                 timeout_s=600,