[CI] skip vllm_example ray-project#36665

Why are these changes needed?
vllm requires cuda to be build, which is not available in our CI gpu docker image.



No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'
--
  | Traceback (most recent call last):
  | File "/opt/miniconda/lib/python3.8/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in <module>
  | main()
  | File "/opt/miniconda/lib/python3.8/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
  | json_out['return_val'] = hook(**hook_input['kwargs'])
  | File "/opt/miniconda/lib/python3.8/site-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 118, in get_requires_for_build_wheel
  | return hook(config_settings)
  | File "/tmp/pip-build-env-b4hl6g1_/overlay/lib/python3.8/site-packages/setuptools/build_meta.py", line 341, in get_requires_for_build_wheel
  | return self._get_build_requires(config_settings, requirements=['wheel'])
  | File "/tmp/pip-build-env-b4hl6g1_/overlay/lib/python3.8/site-packages/setuptools/build_meta.py", line 323, in _get_build_requires
  | self.run_setup()
  | File "/tmp/pip-build-env-b4hl6g1_/overlay/lib/python3.8/site-packages/setuptools/build_meta.py", line 338, in run_setup
  | exec(code, locals())
  | File "<string>", line 24, in <module>
  | RuntimeError: Cannot find CUDA at CUDA_HOME: /usr/local/cuda. CUDA must be available in order to build the package.


Also there is a bug in the build rule cc983fc#r119162333

.buildkite/pipeline.gpu_large.yml

@@ -50,7 +50,7 @@
  ["NO_WHEELS_REQUIRED", "RAY_CI_PYTHON_AFFECTED", "RAY_CI_TUNE_AFFECTED", "RAY_CI_DOC_AFFECTED"]
  commands:
  - cleanup() { if [ "${BUILDKITE_PULL_REQUEST}" = "false" ]; then ./ci/build/upload_build_info.sh; fi }; trap cleanup EXIT
- - DOC_TESTING=1 TRAIN_TESTING=1 TUNE_TESTING=1 INSTALL_VLLM=1 ./ci/env/install-dependencies.sh
+ - DOC_TESTING=1 TRAIN_TESTING=1 TUNE_TESTING=1 ./ci/env/install-dependencies.sh
  - pip install -Ur ./python/requirements/ml/requirements_ml_docker.txt
  - ./ci/env/env_info.sh
  # Test examples with newer version of `transformers`

ci/env/install-dependencies.sh

@@ -394,11 +394,6 @@ install_pip_packages() {
  requirements_packages+=("holidays==0.24") # holidays 0.25 causes `import prophet` to fail.
  fi
 
- # Additional dependency for vllm.
- if [ "${INSTALL_VLLM-}" = 1 ]; then
- requirements_packages+=("vllm")
- fi
-
  # Data processing test dependencies.
  if [ "${DATA_PROCESSING_TESTING-}" = 1 ] || [ "${DOC_TESTING-}" = 1 ]; then
  requirements_files+=("${WORKSPACE_DIR}/python/requirements/data_processing/requirements.txt")

doc/BUILD

@@ -280,15 +280,6 @@ py_test_run_all_subdirectory(
  tags = ["exclusive", "team:ml", "ray_air", "gpu"],
 )
 
-py_test(
- name = "vllm_example",
- size = "large",
- include = ["source/serve/doc_code/vllm_example.py"],
- exclude = [],
- extra_srcs = [],
- tags = ["exclusive", "team:serve", "gpu"],
-)
-
 py_test(
  name = "pytorch_resnet_finetune",
  size = "large",