Keep track and log the number of daemon starts and restarts

[hangeza]

The daemon start/stop/restart is controlled by the systems systemctl service which automatically restarts the daemon e.g. when this crashes. In order to get an understanding of the reliability of the daemon, it would be useful to track the number of (unintentional by the user) restarts in the detector log to eventually come up with a quantity characterizing the stability of the daemon, such as mean time between failures (MTBF).

To do so, a daemon restart has to be detected externally to the daemon or by the daemon itself through a static restart counter (eg. saved to file) which will be written to the log (MQTT and otherwise) e.g. during the daily log cycle.

This feature would improve the quality and stability of the daemon greatly, since currently there is no quantity whatsoever to estimate the MTBF rate and therefore no countermeasures can be taken for potentially undefined behavior of the daemon.