Add vLLM v0.18.1 source tree with KV transfer abort fix

third_party/vllm/ now tracked in git for direct patch management.
Based on vLLM v0.18.1 release with one patch applied:

  vllm/v1/core/sched/scheduler.py:
    Replace fatal assert with graceful skip when KV transfer callback
    arrives for an already-aborted request during PD disaggregated serving.

Future vLLM modifications should be made directly in third_party/vllm/
and committed normally. The patches/ directory is kept as documentation
of what changed from upstream.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

third_party/vllm/tests/models/language/generation/conftest.py

@@ -0,0 +1,46 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+"""Pytest configuration for vLLM language generation tests."""
+
+import os
+import warnings
+
+import torch
+
+from vllm.platforms import current_platform
+
+
+def pytest_configure(config):
+    """Early ROCm configuration that must happen before test collection."""
+    if not current_platform.is_rocm():
+        return
+
+    # Disable skinny GEMM on ROCm to avoid non-deterministic results
+    # from atomic reductions in wvSplitKrc kernel.
+    # See: https://github.com/vllm-project/vllm/pull/33493#issuecomment-3906083975
+    os.environ["VLLM_ROCM_USE_SKINNY_GEMM"] = "0"
+    warnings.warn(
+        "ROCm: Set VLLM_ROCM_USE_SKINNY_GEMM=0 to avoid non-deterministic "
+        "results from skinny GEMM atomic reductions",
+        UserWarning,
+        stacklevel=1,
+    )
+
+
+def pytest_sessionstart(session):
+    """Configure ROCm-specific settings before test session starts."""
+    if not current_platform.is_rocm():
+        return
+
+    # Disable Flash/MemEfficient SDP on ROCm to avoid HF Transformers
+    # accuracy issues: https://github.com/vllm-project/vllm/issues/30167
+    # TODO: Remove once ROCm SDP accuracy issues are resolved on HuggingFace
+    torch.backends.cuda.enable_flash_sdp(False)
+    torch.backends.cuda.enable_mem_efficient_sdp(False)
+    torch.backends.cuda.enable_math_sdp(True)
+    warnings.warn(
+        "ROCm: Disabled flash_sdp and mem_efficient_sdp, enabled math_sdp "
+        "to avoid HuggingFace Transformers accuracy issues",
+        UserWarning,
+        stacklevel=1,
+    )