fix(observability): 🐛 harden vmagent liveness probe failureThreshold 10→3

Silent outage for 72h went undetected due to lenient probe.
Add startupProbe (failureThreshold=30) to allow slow starts.

otc/dev.t09.de/stacks/observability-client/vm-client-stack/values.yaml

@@ -801,6 +801,20 @@ vmagent:
       # Do not store original labels in vmagent's memory by default. This reduces the amount of memory used by vmagent
       # but makes vmagent debugging UI less informative. See: https://docs.victoriametrics.com/vmagent/#relabel-debug
       promscrape.dropOriginalLabels: "true"
+    # Harden liveness probe: default failureThreshold=10 masked a 72h silent outage
+    livenessProbe:
+      httpGet:
+        path: /health
+        port: http
+      failureThreshold: 3
+      periodSeconds: 5
+      timeoutSeconds: 5
+    startupProbe:
+      httpGet:
+        path: /health
+        port: http
+      failureThreshold: 30
+      periodSeconds: 5
   # -- (object) VMAgent ingress configuration
   ingress:
     enabled: false