The `prometheus-frontend` role of the example app kubeprometheus section
is scoped to the default namespace. Thereby the frontend Prometheus
instance is not able to discover anything outside of the default
namespace. We might as well restrict the front end service monitor to
the default namespace too.

One of the `sed` commands for updating the kubeadm systemd
file (`/etc/systemd/system/kubelet.service.d/10-kubeadm.conf`)
was missing a `-i` reference to the file itself, causing it
to hang indefinitely if ran as declared.

I also wrapped this second `sed` in an `if grep ...` check,
in order to make the operation idempotent.

added service account, role and role binding for the prometheus frontend
example, also updated prometheus to use the correct service account

fixes #1049

Update alert rule for kubelet

Update alert rule check kubelet down ratio from 1% to 10%.
In #774 , it is changed to 1%, so returns to 10%.

Added information on webhook token authentication and information on how to update the manifests if Kubernetes is already deployed with kubeadm.

Add alert if it samples aren't ingested

Update to grafana 5.0.0-beta5

Add a missing resource to the operator's cluster role allowing it to correctly
manager alertmanagers.

Add a missing 'prometheuses/finalizers' resource to the operator cluster role.
Without this role, the operator will fail to create a default secret for a
Prometheus instance in the absence of any service monitors.

The fix seems to already be present in the `examples` version of the role; this
patch brings the contrib version in line with the example.

kube-prometheus: Upgrade to grafana v5

fixing#972

Might as well, right?

As mentioned in https://github.com/coreos/prometheus-operator/issues/924#issuecomment-362859346 .

fixing#947

Promnetheus -> Prometheus

Add role rules permissions to get metrics from kubelet