2.0: Cluster details page doesn't look to always shows correct box status

[eldridgeh]

Summary of Issue: When something happens to a box on a cloud provider (AWS), the state of that box is not reported correctly in the dashboard.

Steps to Reproduce:

1. Spin up a cluster on AWS and go into the cloud providers console and stop the instance.
2. Reload the Clusters details page.

Expected Results: The cluster details page reflects the actual status of the box.

Actual Results: The clusters details page always shows masters and nodes as working

Dev Info: (fill out and add links to log files)

1. go version: 1.10
2. SG latest commit hash or release tag: dddedff38267d34e16763fde1955ca4dec1cc5d7
3. Number of Masters and Nodes: Any
4. cloud provider: Any

[eldridgeh]

@gopherstein This is still an issue. After stopping the instance in AWS the dashboard never showed that there was any change to the node I stopped. After reloading the UI, it still showed the stopped node as active. Metrics would not pull either. Once I started the node back up it joined the cluster and metrics started pulling again. In short, we need to display actual box statuses in the UI.