Verification / Detection follow-ups

[RobertJoonas]

Changes

1. Consider internal check timeouts to be Browserless issues in both detection and verification
2. Assert on captured sentry messages

Note: This PR makes the timeout_ms (that is currently fixed for each InstallationSupport.Check module) configurable on the CheckRunner level. This will also be useful for increasing detection timeout for the domain change flow.

Tests

• Automated tests have been added

Changelog

• This PR does not make a user-facing change

Documentation

• This change does not need a documentation update

Dark mode

• This PR does not change the UI

[apata]

suggestion, non-blocking: checks_test, checks_observability_test for both Verification and Detection have quite a bit of similar code, like stubbing, test sentry conf, run_checks, test scenarios.

I think we could cut down on this duplication a lot by doing the following:
In the main checks_test modules, we make sure that expected %Result's are achieved.
In o11y-related unit tests, we make sure that given certain %Result's, handle_result triggers certain logs, metrics etc

[RobertJoonas]

Fair suggestion by all means, but I personally like how these tests do not worry about any intermediate results and focus solely on how a certain Browserless response gets handled. They're still fast, cheap and easy to follow.