[CI] Multi-Node CI scalable (#3611)

### What this PR does / why we need it?
This PR adds a jinja template for the k8s configuration file, prepare
for the upcoming 4-node CI
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: wangli <wangli858794774@gmail.com>

.github/workflows/multi_node_test.yaml

@@ -35,6 +35,7 @@ jobs:
            # configure apt and pip source
            sed -i 's|ports.ubuntu.com|mirrors.tuna.tsinghua.edu.cn|g' /etc/apt/sources.list
            pip config set global.index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
+           pip install jinja2-cli -y
 
            apt-get update -y && apt-get install -y git curl
 
@@ -58,7 +59,13 @@ jobs:
 
         - name: Launch cluster
           run: |
-            kubectl apply -f tests/e2e/multi_node/scripts/lws.yaml
+            jinja2 tests/e2e/multi_node/scripts/lws.yaml.jinja2 \
+              -D size=2 \
+              -D replicas=1 \
+              -D image="m.daocloud.io/quay.io/ascend/cann:8.2.rc1-a3-ubuntu22.04-py3.11" \
+              --outfile lws.yaml
+
+            kubectl apply -f ./lws.yaml
           
         - name: Waiting for pod ready
           run: |
@@ -115,4 +122,4 @@ jobs:
           if: always()
           run: |
             kubectl get pods -n $NAMESPACE
-            kubectl delete -f tests/e2e/multi_node/scripts/lws.yaml
+            kubectl delete -f ./lws.yaml

.github/workflows/vllm_ascend_test_nightly.yaml

@@ -26,9 +26,7 @@ on:
     branches:
       - 'main'
       - '*-dev'
-    paths:
+    types: [labeled]
-      - 'tests/e2e/nightly/**'
-      - '.github/workflows/vllm_ascend_test_nightly.yaml'
 
 # Bash shells do not use ~/.profile or ~/.bashrc so these shells need to be explicitly
 # declared as "shell: bash -el {0}" on steps that need to be properly activated.
@@ -45,6 +43,7 @@ concurrency:
 
 jobs:
   qwen3-32b:
+    if: contains(github.event.pull_request.labels.*.name, 'run-nightly')
     strategy:
       matrix:
         # should add A3 chip runner when available
@@ -57,6 +56,7 @@ jobs:
       runner: ${{ matrix.os }}
       tests: tests/e2e/nightly/models/test_qwen3_32b.py
   qwen3-32b-in8-a3:
+    if: contains(github.event.pull_request.labels.*.name, 'run-nightly')
     strategy:
       matrix:
         os: [ linux-aarch64-a3-4 ]
@@ -67,6 +67,7 @@ jobs:
       image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.2.rc1-a3-ubuntu22.04-py3.11
       tests: tests/e2e/nightly/models/test_qwen3_32b_int8.py
   qwen3-32b-in8-a2:
+    if: contains(github.event.pull_request.labels.*.name, 'run-nightly')
     strategy:
       matrix:
         os: [ linux-aarch64-a2-4 ]
@@ -76,6 +77,7 @@ jobs:
       runner: ${{ matrix.os }}
       tests: tests/e2e/nightly/models/test_qwen3_32b_int8.py
   qwen3-235b-a22b-w8a8-eplb:
+    if: contains(github.event.pull_request.labels.*.name, 'run-nightly')
     strategy:
       matrix:
         # should add A3 chip runner when available
@@ -89,6 +91,7 @@ jobs:
       image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.2.rc1-a3-ubuntu22.04-py3.11
       tests: tests/e2e/nightly/models/test_qwen3_235b_a22b_w8a8_eplb.py
   deepseek-r1-w8a8-eplb:
+    if: contains(github.event.pull_request.labels.*.name, 'run-nightly')
     strategy:
       matrix:
         # should add A3 chip runner when available

tests/e2e/nightly/multi_node/scripts/lws.yaml.jinja2

@@ -4,9 +4,9 @@ metadata:
   name: vllm
   namespace: vllm-project
 spec:
-  replicas: 1
+  replicas: {{ replicas | default(1) }}
   leaderWorkerTemplate:
-    size: 2
+    size: {{ size | default(2) }}
     restartPolicy: RecreateGroupOnPodRestart
     leaderTemplate:
       metadata:
@@ -15,7 +15,7 @@ spec:
       spec:
         containers:
           - name: vllm-leader
-            image: m.daocloud.io/quay.io/ascend/cann:8.2.rc1-a3-ubuntu22.04-py3.11
+            image: {{ image | default("m.daocloud.io/quay.io/ascend/cann:8.2.rc1-a3-ubuntu22.04-py3.11") }}
             env:
               - name: WORKSPACE
                 value: "/root/workspace"
@@ -24,8 +24,6 @@ spec:
                 value: "v0.11.0"
               - name: VLLM_ASCEND_VERSION
                 value: "main"
-              - name: MOONCAKE_VERSION
-                value: "06cc217504a6f1b0cdaa26b096b985651b262748"
             command:
               - sh
               - -c
@@ -70,7 +68,7 @@ spec:
       spec:
         containers:
           - name: vllm-worker
-            image: m.daocloud.io/quay.io/ascend/cann:8.2.rc1-a3-ubuntu22.04-py3.11
+            image: {{ image | default("m.daocloud.io/quay.io/ascend/cann:8.2.rc1-a3-ubuntu22.04-py3.11") }}
             env:
               - name: WORKSPACE
                 value: "/root/workspace"
@@ -79,8 +77,6 @@ spec:
                 value: "v0.11.0"
               - name: VLLM_ASCEND_VERSION
                 value: "main"
-              - name: MOONCAKE_VERSION
-                value: "06cc217504a6f1b0cdaa26b096b985651b262748"
             command:
               - sh
               - -c