Add scrapeTimeout to cluster monitoring config

Current Cluster CR spec allows to set custom interval for ServiceMonitor: https://github.com/rook/rook/blob/09bbdd1c16d8e73b08f28380bf29d20f4d69d1f7/pkg/apis/ceph.rook.io/v1/types.go#L618-L620 but there is no way to set custom scrapeTimeout . If ServiceMonitor spec is updated manually but user, then Rook reverts the change on the next reconcile.

Proposal:

Add optional scrapeTimeout to Cluster spec.montoring. When set, adds scrapeTimeout to ServiceMonitor endpoint.

 apiVersion: ceph.rook.io/v1
 kind: CephCluster
 spec:
   ...
   monitoring:
     enabled: false
     metricsDisabled: false
     interval: 60s
+    scrapeTimeout: 30s
     ...