[CI][Intel GPU] refine intel GPU ci docker build (#14860)

Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>

.buildkite/run-xpu-test.sh

@@ -4,16 +4,27 @@
 # It serves a sanity check for compilation and basic model usage.
 set -ex
 
+image_name="xpu/vllm-ci:${BUILDKITE_COMMIT}"
+container_name="xpu_${BUILDKITE_COMMIT}_$(tr -dc A-Za-z0-9 < /dev/urandom | head -c 10; echo)"
+
 # Try building the docker image
-docker build -t xpu-test -f Dockerfile.xpu .
+docker build -t ${image_name} -f Dockerfile.xpu .
 
 # Setup cleanup
-remove_docker_container() { docker rm -f xpu-test || true; }
+remove_docker_container() { 
+  docker rm -f "${container_name}" || docker image rm -f "${image_name}" || true;
+}
 trap remove_docker_container EXIT
 remove_docker_container
 
 # Run the image and test offline inference/tensor parallel
-docker run --name xpu-test --device /dev/dri -v /dev/dri/by-path:/dev/dri/by-path --entrypoint="" xpu-test sh -c '
+docker run \
+    --device /dev/dri \
+    -v /dev/dri/by-path:/dev/dri/by-path \
+    --entrypoint="" \
+    --name "${container_name}" \
+    "${image_name}" \
+    sh -c '
     python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m
     python3 examples/offline_inference/basic/generate.py --model facebook/opt-125m -tp 2
 '