Create a grafana dashboard for the index-observer service

[davidd8]

Much like storetheindex, the index-observer should have a dedicated grafana dashboard that tracks:

• Operational metrics, such as health checks, uptime, machine stats.
• Business metrics, such as the overall index deal coverage, number of SP index providers, etc.

This will help monitor and maintain the service, as well as ensure stability of the metrics reported for SitReps.

[davidd8]

cc @masih

[masih]

Operational metrics should move to a dashboard dedicated to the K8S cluster. Because, we can then set it up once and benefit from it for all pods and services deployed on the cluster, labeled by service. This will reduce duplicate dashboard set up and make it easier to make a judgement on how the cluster is doing from operational perspective. See ipni/storetheindex#840