Implement health status features

[davidverweij]

API health/status endpoints, which later will be transformed into periodic status emails?

• Monitor failed upload groups with identifiers for resolving

When multiple files are downloaded for a single DMP upload, one of the files might error, effectively cancelling the whole DMP upload. We want to stay informed of these cases, even more so when they persist.

• Total number of unresolved records (records with issues), overall and per technology

A high-level counter to indicate how many records (== data files) have issues (== not uploaded to DMP). Nice to have: elaborated details on 'where' it fails (e.g., no device_id, or no patient_id).

• Think through scenarios of possible failures (e.g., human error on UCAM, or elsewhere)

• Authentication errors with 3rd party services (e.g. Dreem login)

• Do we want to offer the ability to 'ignore' faulty records?

For the benefit of data collection, possible 'quick fixes' by ignoring files and continuing upload for the rest (see also group identifier/download faults). Then, in a non-destructive way and being able to revert?