What would you like to be added:
Cluster-autoscaler currently exposed a lot of metrics that are being scraped by prometheus
We need to leverage these metrics and build meaningful dashboards that can provide useful information to users and developers. We need to do the following

1. Build a standalone cluster-autoscaler dashboard highlighting all key metrics
2. Integrate metrics from cluster-autoscaler with dashboards of machine-controller-manager. This will help show their dependency and how these two actors influence each other in real time

Why is this needed:
This will help improve dev/ops productivity. These dashboards will also help in investigations and analysis