Collect internal metrics of the configuration server and expose them via an Prometheus endpoint

[mariusoe]

In this ticket, the configuration server should be extended that metrics of its internal functionallity can be collected and exposed, thus, it can be queried and stored in a database in order to monitor it.

The metrics should include:

• statistics of the provided Rest endpoints
• active users (valid tokens?)
• count of configuration files and agent mappings
• how often configurations are fetched
• JVM statistics

It should be considered and evaluated whether the Ocelot agent itself could be used for this kind of monitoring.

See #353