[XPU CI] Pull prebuilt nightly image instead of building per-stage (#27282)

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.github/workflows/pr-test-xpu.yml

@@ -73,57 +73,52 @@ jobs:
     needs: [check-changes, pr-gate]
     if: needs.check-changes.outputs.main_package == 'true'
     runs-on: intel-bmg
+    env:
+      DOCKERHUB_INTEL_USERNAME: ${{ secrets.DOCKERHUB_INTEL_USERNAME }}
+      DOCKERHUB_INTEL_TOKEN: ${{ secrets.DOCKERHUB_INTEL_TOKEN }}
     steps:
+      - name: Reset workspace ownership
+        run: |
+          # Prior runs install sglang as root inside the CI container against
+          # the bind-mounted workspace, leaving root-owned egg-info on the host.
+          # actions/checkout's pre-clean step can't unlink those as the runner
+          # user, so reset ownership before checkout runs.
+          docker run --rm -v "${{ github.workspace }}:/w" busybox:latest \
+            chown -R "$(id -u):$(id -g)" /w || true
+
       - name: Checkout code
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
           ref: ${{ inputs.ref || github.ref }}
 
-      - name: Build Docker image
+      - name: Start CI container
         run: |
-          PR_REPO=${{ github.event.pull_request.head.repo.clone_url }}
-          PR_HEAD_REF=${{ github.head_ref }}
-          docker build \
-            ${PR_REPO:+--build-arg SG_LANG_REPO=$PR_REPO} \
-            ${PR_HEAD_REF:+--build-arg SG_LANG_BRANCH=$PR_HEAD_REF} \
-            --no-cache --progress=plain -f docker/xpu.Dockerfile -t xpu_sglang_main:bmg .
-
-      - name: Run container
-        id: start_container
-        run: |
-          container_id=$(docker run -dt \
-            --group-add 992 \
-            --group-add $(getent group video | cut -d: -f3) \
-            --group-add $(getent group render | cut -d: -f3) \
-            -v $HOME/.cache/huggingface:/root/.cache/huggingface \
-            --device /dev/dri \
-            -v /dev/dri/by-path:/dev/dri/by-path \
-            -e HF_TOKEN="$(cat ~/huggingface_token.txt)" \
-            xpu_sglang_main:bmg)
-          echo "Started container: $container_id"
-          echo "container_id=$container_id" >> "$GITHUB_OUTPUT"
+          export HF_TOKEN="$(cat ~/huggingface_token.txt)"
+          bash scripts/ci/xpu/xpu_ci_start_container.sh
+        env:
+          GITHUB_WORKSPACE: ${{ github.workspace }}
 
       - name: Install Dependency
-        timeout-minutes: 20
+        timeout-minutes: 60
         run: |
-          cid="${{ steps.start_container.outputs.container_id }}"
-          docker exec "$cid" /opt/venv/bin/python3 -m pip install --upgrade pip
-          docker exec "$cid" /opt/venv/bin/python3 -m pip install pytest expecttest ray huggingface_hub tabulate
-          docker exec "$cid" /opt/venv/bin/python3 -m pip uninstall -y flashinfer-python
-          docker exec "$cid" /bin/bash -c '/opt/venv/bin/hf auth login --token ${HF_TOKEN} '
+          docker exec ci_sglang_xpu /opt/venv/bin/python3 -m pip install --upgrade pip
+          docker exec ci_sglang_xpu /opt/venv/bin/python3 -m pip install pytest expecttest ray huggingface_hub tabulate
+          docker exec ci_sglang_xpu /opt/venv/bin/python3 -m pip uninstall -y flashinfer-python sgl-kernel sglang
+          docker exec ci_sglang_xpu cp /sglang-checkout/python/pyproject_xpu.toml /sglang-checkout/python/pyproject.toml
+          docker exec -w /sglang-checkout/python ci_sglang_xpu /opt/venv/bin/python3 -m pip install --no-cache-dir . --extra-index-url https://download.pytorch.org/whl/xpu
+          docker exec ci_sglang_xpu /opt/venv/bin/python3 -m pip install --no-cache-dir --no-deps xgrammar==0.1.33
+          docker exec ci_sglang_xpu /opt/venv/bin/python3 -m pip install triton-xpu==3.7.1 --index-url https://download.pytorch.org/whl/test/xpu --force-reinstall
+          docker exec ci_sglang_xpu /bin/bash -c '/opt/venv/bin/hf auth login --token ${HF_TOKEN}'
 
       - name: Run stage-a tests
         timeout-minutes: 30
         run: |
-          cid="${{ steps.start_container.outputs.container_id }}"
-          docker exec "$cid" bash -c "source /opt/venv/bin/activate && cd /sgl-workspace/sglang/test && python3 run_suite.py --hw xpu --suite stage-a-test-1-gpu-xpu"
+          docker exec ci_sglang_xpu bash -c "source /opt/venv/bin/activate && cd /sglang-checkout/test && python3 run_suite.py --hw xpu --suite stage-a-test-1-gpu-xpu"
 
       - name: Cleanup container
         if: always()
-        run: |
-          cid="${{ steps.start_container.outputs.container_id }}"
-          docker rm -f "$cid" || true
+        run: docker rm -f ci_sglang_xpu || true
 
   # ==================== Wait for Stage A ==================== #
   wait-for-stage-a: