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/requirements/xpu.txt

@@ -0,0 +1,18 @@
+# Common dependencies
+-r common.txt
+
+ray>=2.9
+cmake>=3.26.1
+packaging>=24.2
+setuptools-scm>=8
+setuptools>=77.0.3,<81.0.0
+wheel
+jinja2>=3.1.6
+datasets # for benchmark scripts
+numba == 0.61.2 # Required for N-gram speculative decoding
+--extra-index-url=https://download.pytorch.org/whl/xpu
+torch==2.10.0+xpu
+torchaudio
+torchvision
+
+vllm_xpu_kernels @ https://github.com/vllm-project/vllm-xpu-kernels/releases/download/v0.1.3/vllm_xpu_kernels-0.1.3-cp38-abi3-linux_x86_64.whl