Improve Cassandra deployment for Kubernetes / OpenShift

[jpkrohling]

During the initial Docker/Kubernetes/OpenShift , we took some shortcuts regarding Cassandra and they need to be sorted out.

Points to solve:

• Use a more appropriate Cassandra Docker image
• Wrap the Docker entry point to run a nodetool drain on shutdown, as suggested by @jsanda
• Figure out the best approach regarding the replication strategy and datacenter configuration

[pavolloffay]

fixed issue for replication strategy jaegertracing/jaeger-kubernetes#15

[pavolloffay]

nodetool drain in hawkular-metrics

https://github.com/openshift/origin-metrics/blob/master/cassandra/cassandra-prestop.sh
https://github.com/openshift/openshift-ansible/blob/master/roles/openshift_metrics/templates/hawkular_cassandra_rc.j2#L114-L123

[pavolloffay]

@jpkrohling I think this can be closed

nodetool drain jaegertracing/jaeger-kubernetes#21 and jaegertracing/jaeger-openshift#27

Configuration for replication factor is already in master.