feat: add health check for epp cluster

Signed-off-by: zhengkezhou1 <madzhou1@gmail.com>

Author: zhengkezhou1

test/testdata/envoy.yaml

@@ -170,6 +170,14 @@ data:
                 max_pending_requests: 40000
                 max_requests: 40000
                 max_retries: 1024
+          health_checks:
+            - timeout: 1s
+              interval: 5s
+              unhealthy_threshold: 3
+              healthy_threshold: 1
+              grpc_health_check:
+                service_name: "inference-extension"
+                authority: vllm-llama3-8b-instruct-epp.$E2E_NS:9003
           # This ensures that envoy accepts untrusted certificates. We tried to explicitly
           # set TrustChainVerification to ACCEPT_UNSTRUSTED, but that actually didn't work
           # and what worked is setting the common_tls_context to empty.                
@@ -193,6 +201,8 @@ data:
                   region: ext_proc/e2e/0
                 lb_endpoints:
                   - endpoint:
+                      health_check_config:
+                        port_value: 9003
                       address:
                         socket_address:
                           address: vllm-llama3-8b-instruct-epp.$E2E_NS

test/testdata/inferencepool-e2e.yaml

@@ -20,10 +20,15 @@ spec:
   selector:
     app: vllm-llama3-8b-instruct-epp
   ports:
-    - protocol: TCP
+    - name: http2-por
+      protocol: TCP
       port: 9002
       targetPort: 9002
       appProtocol: http2
+    - name: health-check-port
+      protocol: TCP
+      port: 9003
+      targetPort: 9003
   type: ClusterIP
 ---
 apiVersion: apps/v1
@@ -62,6 +67,8 @@ spec:
         - "9002"
         - -grpcHealthPort
         - "9003"
+        - -secureServing
+        - "false"
         env:
         - name: USE_STREAMING
           value: "true"
@@ -109,7 +116,7 @@ rules:
   - subjectaccessreviews
   verbs:
   - create
---- 
+---
 kind: ClusterRoleBinding
 apiVersion: rbac.authorization.k8s.io/v1
 metadata: