Debugging Logs and Host Container Status using Control Container

This issue is catalyzed by this experience: https://github.com/bottlerocket-os/bottlerocket/discussions/1573

In the case that the admin container is not starting or unusable, there are no good facilities for
- Knowing the state of the admin container (or any host container)
- Debugging the cause of host container failure
- Debugging Bottlerocket, in general, without using sudo sheltie in the admin container.

**What I'd like:**

Improved ability to debug Bottlerocket problems without using the admin container.

**Any alternatives you've considered:**

Here some possibilities that come to mind, some better/simpler than others:
- Providing a route in the API that calls logdog and returns these logs to the caller. `*`
- API routes that give host container status and logs from host containers.
- An apiclient call for starting or changing/restarting a host-container that blocks and returns the outcome.
- API calls for getting systemd unit statuses

`*` note that logdog does not currently include specific host-container calls such as `systemctl status admin@host-container.service`