Have a resource-usage summary dashboard

It would be great to see how many resources (pod count, memory usage, cpu usage, etc.) are in use for Sematic runs. The natural workaround for this, which doesn't require any Sematic features, would be to use standard monitoring tools and look at these quantities for pods in the namespace where Sematic is deployed.