fix(docker): make docker more robust towards pausing,restarts, missing healtchecks and missing container states

[CommanderStorm]

_{ℹ️ To keep reviews fast and effective, please make sure you’ve read our pull request guidelines}

📝 Summary of changes done and why they are done

Tries to fix #6292 (comment)

@mayberryjp @benji2k2 @bondskin could you test this PR?
Instructions for how to test PRs are here:

https://github.com/louislam/uptime-kuma/wiki/Test-Pull-Requests

📋 Related issues

• Relates to fix(docker): handle the state "unhealthy" as DOWN instead of PENDING #6292 (comment)

[benji2k2]

In general the changes should work but they change the behaviour in some cases. In example containers without a health-state stay „PENDING“ instead of being „UP“ like before. In my opinion the most „minimal invasive“ solution solving the original problem with „unhealthy“ containers was my original pr (31eaf2d). This should only solve the original problem and does not have any side-effects in other cases (it is running here on 3 servers without any problems for three months). Maybe you think about using this solution to avoid other problems.

[benji2k2]

@CommanderStorm This should restore the old behavior for containers without health state. I tested the PR and it is working for running, paused, healthy and unhealthy containers. I cannot test it for the case without health-status and for restarting containers in my setup. but looks good so far.

[CommanderStorm]

Works for me, lets merge

[CommanderStorm]

FYI: If we migrate this monitor to the newer monitor condition way of building them, this would enable writing testcases for this monitor and prevent issues like this