Print all test case error messages together at the end

[wilfonba]

Description

I have recently found the behavior described in #744 to be annoying. This PR addresses that by keeping a list of all failed tests and printing the errors all together at the end.

Fixes #744

Type of change

Please delete options that are not relevant.

• New feature (non-breaking change which adds functionality)

Scope

• This PR comprises a set of related changes with a common goal

How Has This Been Tested?

Test 1: I intentionally caused tests to fail due to tolerance by commenting out dvd(0) in the fifth order WENO reconstruction so the solution would be wrong. The resulting output of ./mfc.sh test -j 1 -o bc=-1 > test1.txt (without coloring) is
test1.txt

Test 2: Intentionally caused tests to fail due to NaNs by commenting out beta(0) in the fifth-order WENO reconstruction. The resulting output of ./mfc.sh test -j 1 -o bc=-1 > test2.txt (without color) is
test2.txt

Test 3: Intentionally caused tolerance errors in the WENO routine and skipped some tests. The result of running ./mfc.sh test -j 8 -o 1D -% 80 > test4.txt (without color) is
test4.txt

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 43.63%. Comparing base (a3d39df) to head (ab90416).
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #745   +/-   ##
=======================================
  Coverage   43.63%   43.63%           
=======================================
  Files          61       61           
  Lines       16893    16893           
  Branches     1948     1948           
=======================================
  Hits         7372     7372           
  Misses       8404     8404           
  Partials     1117     1117           

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

[sbryngelson]

@henryleberre might comment on the 'cleanness' of the toolchain implementation. For me, it seems reasonable and gets the job done.

[sbryngelson]

Is this in conflict with something added by @henryleberre in #737 ?

[wilfonba]

Merged without issue

[sbryngelson]

@wilfonba #737 is still open (not merged)

[wilfonba]

@wilfonba #737 is still open (not merged)

Oh. Yeah, it probably does conflict in some way with those changes, although minor. This can wait for #737 to be closed. I can't imagine much else causing conflicts once merged.