Handle active Sections for fatal errors

[hbirler]

Closes #1210

When a signal is caught, the destructors of Sections will not be called. Thus, we must call sectionEndedEarly manually for those Sections.

Example test case:

TEST_CASE("broken") {
   SECTION("section") {
      /// Use illegal cpu instruction
      __asm__ __volatile__("ud2" : : : "memory");
   }
}

[hbirler]

I am not familiar with the internals of Catch2, so there might be a nicer way of doing this. For example, I am not sure whether it is possible to properly construct SectionEndInfo endInfo with correct prevAssertions and durationInSeconds

[hbirler]

Additionally, it might be safe to do:

auto nl = CATCH_MOVE(m_activeSections.back()->nameAndLocation());

instead of a copy, but I took the "safe" route (allocating memory might also not be the best idea within a signal handler).

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 5 lines in your changes are missing coverage. Please review.

Project coverage is 91.04%. Comparing base (029fe3b) to head (fe428d6).
Report is 5 commits behind head on devel.

❗ Current head fe428d6 differs from pull request most recent head 0b013eb. Consider uploading reports for the commit 0b013eb to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##            devel    #2855      +/-   ##
==========================================
- Coverage   91.09%   91.04%   -0.05%     
==========================================
  Files         197      198       +1     
  Lines        8382     8391       +9     
==========================================
+ Hits         7635     7639       +4     
- Misses        747      752       +5     

[hbirler]

The function RunContext::handleFatalErrorCondition seems to not be covered at all by the existing tests, and I wouldn't know how to test signal handlers reliably within the existing framework.

[horenmar]

Thanks, this has been around for a while.