Aggregation in the bridge

[norro]

Allow bridge to be configured to aggregate diagnostic status messages before sensing a diagnostic array.
E.g. aggregate for a certain duration (publishing array with fixed rate), or aggregate until status for a defined set of diagnostic tasks was received.

Depends on #26, might become more relevant with #16.

[bjv-capra]

I've also noticed this, what do you think about having a map of diagnostic arrays configured at startup when the lookup table is loaded. AFAIK, the diagnostic array is per updater_id. Thus, we'd need to have a diagnostic array per updater_id and fill it up when the tasks arrive, and as you suggested, wait for X time and send what the map had at that time.