Allow storing job status and logs in a shared location

It would be very interesting in a cluster to be able to access the status and log of jobs run by other nodes. It's also a requirement for XWIKI-22847.

• main metadata: all jobs have standard metadata (start/end date, group, etc.) that would be interesting to be able, like logs, to filter, order and paginate too so the database seems like a good fit
• log: what makes the most sense is probably the database, one raw per log event. Each log event is generally small, and it's very important to be able to filter, order and paginate logs
• custom status: on the other the rest of the status (the custom part) is generally a big black box that you just entirely serialize or unserializer, which is something that is matching more an object storage, like S3