Improve dashboarding in Grafana using provided tools for existing monitors

Grafana is deployed with Prometheus Operator in a totally stateless way, with no persistent storage. Unfortunately, that means that to edit dashboards you have to edit the JSON configmap that defines those dashboards. To make this easier Prometheus Operator provides some Python scripts.

The following parts of the dashboard, so far, need to be fixed:

• Instant mode for all dials and counters
• e2e latency metrics incorrect
• add cpu and memory limits for containers and change fill to fill only current usage on Pod dashboard
• add a graph for temperature of node on Node dashboard, using hwmon metrics
• add a new dashboard for Prometheus-oriented metrics, ideally the default dashboard?
• add a new dashboard for nginx VTS metrics
• Fix bad range for memory usage for pods in Pods
• Increase resolution of node/CPU chart from 1/10 to at least 1/2
• Increase resolution of node/Load 1m chart from 1/4 to 1/1