(WIP) [DPE-5303] Update endpoints faster when snap is stopped

[marceloneppel]

Issue

If the snap services are stopped manually by calling snap stop charmed.postgresql, the endpoints are not updated in the database relation endpoint, causing errors in the client charm side.

Solution

Improve the Cluster Topology Observer to check other members' REST API when the current member's REST API is not accessible.

This will allow the endpoint to be changed to the IP of the new primary before the update-status hook is called the next time.

I'm still working on making this more robust.

[codecov]

Codecov Report

Attention: Patch coverage is 30.30303% with 23 lines in your changes missing coverage. Please review.

Project coverage is 70.60%. Comparing base (01cfd3e) to head (6b02bf1).
Report is 101 commits behind head on main.

Files with missing lines	Patch %	Lines
src/cluster_topology_observer.py	20.00%	20 Missing ⚠️
src/cluster.py	50.00%	3 Missing ⚠️

❌ Your patch check has failed because the patch coverage (30.30%) is below the target coverage (33.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #610      +/-   ##
==========================================
- Coverage   70.81%   70.60%   -0.21%     
==========================================
  Files          12       12              
  Lines        3039     3055      +16     
  Branches      537      542       +5     
==========================================
+ Hits         2152     2157       +5     
- Misses        771      782      +11     
  Partials      116      116              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.