[Postgres] Fix slot recovery & improve log output

[rkistner]

Fixes regression from #150.

When starting replication, there are 4 different scenarios we cater for, depending on whether or not we have already completed the initial snapshot before, and whether or not we have a working logical replication slot.

1. haveSnapshot: false, haveWorkingSlot: false -> initial sync from scratch
2. haveSnapshot: true,  haveWorkingSlot: true  -> resume streaming replication
3. haveSnapshot: false, haveWorkingSlot: true  -> resume initial sync (new in #150)
4. haveSnapshot: true,  haveWorkingSlot: false -> initial sync from scratch (broken in #150)

The regression was in case 4 above - the slot would not get re-created, and replication would remain stuck. It can be triggered by something like replication lag falling too far behind, or anything else that makes the replication slot invalid.

The fix here is to raise a MissingReplicationSlotError error again for that case. That is caught higher up in the process, creates a new replication slot with an incremented index (while keeping the current data active), and then replicating with that new slot.

This PR also makes minor improvements to log output.

[changeset-bot]

🦋 Changeset detected

Latest commit: 7bfecb9

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 6 packages

Name	Type
@powersync/service-module-postgres	Patch
@powersync/service-core	Patch
@powersync/service-image	Patch
@powersync/service-module-mongodb	Patch
@powersync/service-module-mysql	Patch
test-client	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR