[Meta] Fleet Server observability

[mostlyjason]

Observability for Fleet Servers is important because operators are expected to scale them manually. They are also expected to troubleshoot operational issues and fix them. This is true in various degrees from self-managed environments, to ECE/ECK and ESS. As a result, we should provide users visibility to issues and documentation on how to address common ones.

As a Fleet operator, I need:

• Logs, metric and status information to help me scale and troubleshoot issues [Elastic Agent] Logs, metrics and status info for Fleet server beats#24415
• Fleet Server integration dashboard [Fleet Server] Fleet Server Observability integrations#812
• Documentation for Fleet Server operators on to observe it, scale it, scaling limits, restrictions, etc.

Out of scope:

• A notification in the Fleet app when Fleet Server is not healthy so I understand why updates are not being applied [Fleet] Notification when Fleet Server is not healthy kibana#95572

Related issues:

• Agent integration dashboard https://github.com/elastic/beats/issues/23948
• Enable system metrics on Elastic Cloud agent policy [Fleet] Enable system metrics on Elastic Cloud agent policy kibana#96248

[jen-huang]

I merged the requirements here into the main meta issue for Fleet Server GA: #93. Closing.