Container process monitoring

Let's add a monitoring feature of container processes.The superadmin and the user should be able to inspect to what all/their individual containers are doing.

- [x] **agent**: Add a periodic timer to retrieve active kernel container's process list with their basic resource usage such as CPU, memory, IO bytes/sec, etc. Store them in an auto-expiring Redis key.\* The timer may be combined with the existing statistics infra, but the process list should be a live snapshot instead of a time-series.
- We need to exclude some system and krunner processes for brevity.
- Could we combine the GPU stats as well?

- [ ] **manager**: Add a GraphQL field to `ComputeContainer` to fetch this data from Redis.
- [ ] **webui**: Add a UI to the session container list to inspect running processes.
- [ ] **control-panel**: Same to webui.

JIRA Issue: BA-251