Elastic, serverless Portals deployment

[jspenger]

The final goal for the first release is to have an elastic, serverless Portals deployment. For the elastic part, the runtime executes on a Kubernetes cluster, and automatically makes decisions whether to scale up or down. For the serverless part, every node is to be a job manager, and accesses an external RSM for managing the deployments metadata consistently. It should be possible to submit new jobs to the deployment, and the deployment should scale up and down with the demand. There should also be a benchmark to evaluate the performance of this deployment mode. Ask @jonasspenger for more details.