Per issue 593, append .yml to each filename used in the generated configmap. I wasn't able to see alerts in the Prometheus UI until I did this

Update rules

After cloning the repo I wanted to deploy kube-prometheus into my running minikube. I did the steps mentioned in the docu which brought me to the following exception:

namespace "monitoring" created
error: the path "manifests/prometheus-operator" does not exist
Waiting for Operator to register custom resource definitions.................................................................................................................^C

Fixing #751 K8SApiServerLatency always triggering

Metrics servers

* Generated by `make generate` in `contrib/kube-prometheus`

* If any of the rules is active > 10m
* If all daemonsets are not ready
* If all daemonsets are not scheduled
* If some are miss scheduled

kube-prometheus: add stateful sets dashboard

Convert Grafana dashboards to grafanalib scripts

Omit static compute resource requests and limits from deployment

The AWS CLI SG commands are not necessary, kops creates an SG with these ports open to the correct SG by default

Do not specify static resource settings in the kubernetes-deployment to avoid
unnecessary replacement of pods when the manifest is reapplied (`kubectl
apply`).

The addon-resizer will dynamically set the pod compute resource values. If the
values are also set statically in the deployment configuration, reapplying the
configuration will result in the pods getting replaced. Without the static
resource, the deployment configuration can be reapplied, and the pods will not
be replaced.

This change was also made in the upstream kube-state-metrics example manifests.

https://github.com/kubernetes/kube-state-metrics/pull/285

Fix merging of -datasource.json files

The merge was missing a new line at the end of the merge, so only
the first datasource was being put into the ConfigMap.

In the generated version of this file at /contrib/kube-prometheus/manifests/grafana/grafana-deployment.yaml the version was bumped to 4.5.2 but seems it was overlooked in this templated version of the file.

kube-prometheus: various version updates