Export and scrape metrics from all redis/valkey instances

No redis/valkey instance in the estate exports metrics today. Adds a
pinned redis_exporter (v1.89.0 via the artifactapi dockerhub remote)
next to every instance and two cluster-wide VMPodScrapes so the
observability VMAgent picks them all up.

- enables the valkey-operator exporter sidecar on the jellyfin
  ValkeyCluster, overriding the operator's bare-dockerhub default image
  with the artifactapi-proxied pin
- adds a metrics-exporter sidecar (:9121, port name metrics) to the
  artifactapi/authentik/litellm redis and gitea/netbox valkey
  Deployments, with the monitoring.unkin.net/redis-exporter opt-in pod
  label
- adds VMPodScrape redis-exporters selecting that label in any
  namespace, and VMPodScrape valkey-operator-clusters for
  operator-managed pods whose labels are fixed (the ValkeyCluster CR has
  no pod-label passthrough)

apps/base/netbox/valkey-deployment.yaml

@@ -27,6 +27,7 @@ spec:
         app: netbox-valkey
         app.kubernetes.io/name: netbox
         app.kubernetes.io/component: valkey
+        monitoring.unkin.net/redis-exporter: "true"
     spec:
       securityContext:
         fsGroup: 999
@@ -82,6 +83,26 @@ spec:
           volumeMounts:
             - mountPath: /data
               name: data
+        - name: metrics-exporter
+          image: artifactapi.k8s.syd1.au.unkin.net/dockerhub/oliver006/redis_exporter:v1.89.0
+          imagePullPolicy: IfNotPresent
+          ports:
+            - containerPort: 9121
+              name: metrics
+              protocol: TCP
+          securityContext:
+            allowPrivilegeEscalation: false
+            runAsNonRoot: true
+            capabilities:
+              drop:
+                - ALL
+          resources:
+            limits:
+              cpu: 100m
+              memory: 64Mi
+            requests:
+              cpu: 25m
+              memory: 32Mi
       restartPolicy: Always
       volumes:
         - name: data