[REQUEST]: How to use the inactivity timeout with agent running in kubernetes & containers

### Description

We need to explain to users how managed agents work a containerized environment and guidance on how to configure the agent policy's inactivity timeout to avoid showing many inactive agents. We should also point to guidance on how to adjust resource limits to avoid container restarts.

Any time a container restarts, they get enrolled as a new agent since there's no state saved from the previous enrollment. There is a feature called the "inactive timeout" that can be configured on the agent policy. This will hide the agents when after they restart so they only see the active ones in the UI by default.

If an agent container is restarting very often there's also probably issue that should be investigated to prevent that from happening so often. Most likely their agent pods don't have enough memory or CPU and are hitting their limits, causing k8s to restart the pod.

### Resources

https://www.elastic.co/guide/en/fleet/current/set-inactivity-timeout.html
https://www.elastic.co/guide/en/fleet/current/running-on-kubernetes-managed-by-fleet.html

### Collaboration

The documentation team will investigate the issue and create the initial content.

### Point of contact.

**Main contact:** @

**Stakeholders:**
- @mlunadia
- @nimarezainia
- @michalpristas