Do not block worker threads forever if no event is received (#329)

[frankosterfeld]

Make sure that long-poll request handlers do not block forever when no corresponding event is received. Otherwise the clients will send one request after another once their request times out client-side, until the worker threads are exhausted and the server stops responding.

It would be great if we could also detect the client connection be dropped using Keep-Alive, but cpp-httplib doesn't seem to have API for that.

[codecov[bot]]

Codecov Report

Attention: 12 lines in your changes are missing coverage. Please review.

Comparison is base (c19531d) 55.75% compared to head (260e752) 55.80%.

:exclamation: Current head 260e752 differs from pull request most recent head 6444125. Consider uploading reports for the commit 6444125 to get more accurate results

Files	Patch %	Lines
src/majordomo/include/majordomo/RestBackend.hpp	14.28%	12 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #330 +/- ## ========================================== + Coverage 55.75% 55.80% +0.04% ========================================== Files 69 69 Lines 7291 7299 +8 Branches 2691 2693 +2 ========================================== + Hits 4065 4073 +8 - Misses 1489 1490 +1 + Partials 1737 1736 -1 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[sonarcloud[bot]]

SonarCloud Quality Gate failed.   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

4.9% Coverage
0.0% Duplication

Catch issues before they fail your Quality Gate with our IDE extension SonarLint

[frankosterfeld]

This seems to look OK but should be accompanied by a unit-tests that demonstrate the feature/workaround.

I've made the timeout configurable and also applied it to GET/SET now. I had a hard time to create a working subscription test case, so I've rebased this now against the changes in #331 and made in part of #331 (has both changes now).