[Fleet-Server] Implementation of a Standalone Fleet-Server for Agent Management

[Harmlos]

Describe the enhancement:

A fleet-server should be implemented as a standalone application, launched with a few parameters, including a minimal set of metrics and heartbeat. Additionally, there should be a possibility to launch the fleet-server in a Docker container.

Describe a specific use case for the enhancement or feature:

Currently, to configure the schema elastic-agent -> fleet-server -> Elasticsearch, it is required to deploy an elastic-agent, which occupies 2 GB after installation, with a special policy. After this setup, the agent cannot be used for any other role except as a fleet-server. The fleet-server component itself only occupies 32 MB.

Furthermore, it is not possible to deploy a fleet-server and an agent that runs other integrations on the same host due to port conflicts, which cannot be resolved unless using large Docker images.

Additionally, with a standalone fleet-server, it will be possible to collect events and metrics from the agent using elastic-agent integrations, similar to how integrations with Elasticsearch, Kibana, and Logstash are currently handled.