Improve test suite

[nielstron]

Is your feature request related to a problem? Please describe.

The output of the test system is rather hard to interpret, failing test cases being almost impossible to make out. The python "subtest" system should be introduced as suggested in #162

Further, related to i.e. #157, a class of test cases should be introduced that checks non-occurence of errors rather than the actual output.

Describe the solution you'd like

• Subtest system
• Error catching tests

Describe alternatives you've considered
None

Additional context
#157
#162