cli: reana-cluster init throws an exception if REANA cluster is already running

[hjhsalo]

reana-cluster init throws an exception if REANA cluster already running.
For Kubernetes backend this is an API exception which gets thrown since Kubernetes object(s)(Pod, Namespace, Service, etc.) for a given REANA component already exists after the first initialization.

In order to be more user friendly, this exception should be catched and user should be given a proper explanation about the error.

Even better would be to implement a check to determine if REANA cluster is already initialized, i.e. are there components running in given backend, even before trying to deploy REANA components to the backend.

(cli/cluster.py#L102)
(/backends/kubernetes/k8s.py#L245)

[diegodelemos]

This issue won't be fixed since soon REANA-Cluster will be deprecated in favour of Helm based deployments.