chore: setup for performance benchmarking (#3096)

# What does this PR do?
1. Added a simple mock openai-compat server that serves chat/completion
2. Add a benchmark server in EKS that includes mock inference server
3. Add locust (https://locust.io/) file for load testing

## Test Plan
bash apply.sh
kubectl port-forward service/locust-web-ui 8089:8089
Go to localhost:8089 to start a load test

<img width="1392" height="334" alt="image"
src="https://github.com/user-attachments/assets/d6aa3deb-583a-42ed-889b-751262b8e91c"
/>
<img width="1362" height="881" alt="image"
src="https://github.com/user-attachments/assets/6a28b9b4-05e6-44e2-b504-07e60c12d35e"
/>

docs/source/distributions/k8s-benchmark/locust-k8s.yaml

@@ -0,0 +1,131 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: locust-master
+  labels:
+    app: locust
+    role: master
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: locust
+      role: master
+  template:
+    metadata:
+      labels:
+        app: locust
+        role: master
+    spec:
+      containers:
+      - name: locust-master
+        image: locustio/locust:2.31.8
+        ports:
+        - containerPort: 8089  # Web UI
+        - containerPort: 5557  # Master communication
+        env:
+        - name: LOCUST_HOST
+          value: "${LOCUST_HOST}"
+        - name: LOCUST_LOCUSTFILE
+          value: "/locust/locustfile.py"
+        - name: LOCUST_WEB_HOST
+          value: "0.0.0.0"
+        - name: LOCUST_MASTER
+          value: "true"
+        - name: LOCUST_BASE_PATH
+          value: "${LOCUST_BASE_PATH}"
+        - name: INFERENCE_MODEL
+          value: "${BENCHMARK_INFERENCE_MODEL}"
+        volumeMounts:
+        - name: locust-script
+          mountPath: /locust
+        command: ["locust"]
+        args:
+        - "--master"
+        - "--web-host=0.0.0.0"
+        - "--web-port=8089"
+        - "--host=${LOCUST_HOST}"
+        - "--locustfile=/locust/locustfile.py"
+      volumes:
+      - name: locust-script
+        configMap:
+          name: locust-script
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: locust-worker
+  labels:
+    app: locust
+    role: worker
+spec:
+  replicas: 2  # Start with 2 workers, can be scaled up
+  selector:
+    matchLabels:
+      app: locust
+      role: worker
+  template:
+    metadata:
+      labels:
+        app: locust
+        role: worker
+    spec:
+      containers:
+      - name: locust-worker
+        image: locustio/locust:2.31.8
+        env:
+        - name: LOCUST_HOST
+          value: "${LOCUST_HOST}"
+        - name: LOCUST_LOCUSTFILE
+          value: "/locust/locustfile.py"
+        - name: LOCUST_MASTER_HOST
+          value: "locust-master-service"
+        - name: LOCUST_MASTER_PORT
+          value: "5557"
+        - name: INFERENCE_MODEL
+          value: "${BENCHMARK_INFERENCE_MODEL}"
+        - name: LOCUST_BASE_PATH
+          value: "${LOCUST_BASE_PATH}"
+        volumeMounts:
+        - name: locust-script
+          mountPath: /locust
+        command: ["locust"]
+        args:
+        - "--worker"
+        - "--master-host=locust-master-service"
+        - "--master-port=5557"
+        - "--locustfile=/locust/locustfile.py"
+      volumes:
+      - name: locust-script
+        configMap:
+          name: locust-script
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: locust-master-service
+spec:
+  selector:
+    app: locust
+    role: master
+  ports:
+  - name: web-ui
+    port: 8089
+    targetPort: 8089
+  - name: master-comm
+    port: 5557
+    targetPort: 5557
+  type: ClusterIP
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: locust-web-ui
+spec:
+  selector:
+    app: locust
+    role: master
+  ports:
+  - port: 8089
+    targetPort: 8089
+  type: ClusterIP  # Keep internal, use port-forward to access