chore(longhorn): Extend alert trigger duration

This patch extends the duration for the longhorn-share alert, that was
hotfixed with the hotfix-7183 cronjob. Since the cronjob runs every 5
minutes, this patch extends the alert to 10 minutes to catch possible
hotfix failures.

infrastructure/longhorn/monitoring.yaml

@@ -108,7 +108,7 @@ spec:
         description: Longhorn share manager count is off by {{$value}}. This is likely due to a recent bug in Longhorn. https://github.com/longhorn/longhorn/issues/7183#issuecomment-1823715359
         summary: Longhorn share manager count is off by {{$value}} for 5m.
       expr: count(sum by (namespace, persistentvolumeclaim) (kube_persistentvolumeclaim_access_mode{access_mode="ReadWriteMany"}) * on (namespace, persistentvolumeclaim) group_right kube_persistentvolumeclaim_info{storageclass=~"longhorn.*"}) - sum(kube_pod_info{namespace="longhorn-system", pod=~"share-manager-.*"}) > 0
-      for: 5m
+      for: 10m
       labels:
         issue: Longhorn share manager count is off by {{$value}} for 5m.
         severity: critical