fix: decouple parity violating amplitudes

[spflueger]

Decouple coefficients of amplitudes which are parity violating, by generating the correct coefficient names.

Closes #419

• Add test to verify decoupling for parity violating amplitudes

[spflueger]

@sebastianJaeger Could you verify that this creates the correct amplitude coefficient couplings. I already checked it but just to verify.

[codecov]

Codecov Report

Merging #421 (1474677) into master (d7f81eb) will increase coverage by 0.10%.
The diff coverage is 93.33%.

@@            Coverage Diff             @@
##           master     #421      +/-   ##
==========================================
+ Coverage   89.48%   89.58%   +0.10%     
==========================================
  Files          26       26              
  Lines        3747     3745       -2     
  Branches      926      927       +1     
==========================================
+ Hits         3353     3355       +2     
+ Misses        201      199       -2     
+ Partials      193      191       -2     

Flag	Coverage Δ
unittests	89.58% <93.33%> (+0.10%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/expertsystem/amplitude/helicity_decay.py	90.31% <93.33%> (+1.46%)	⬆️

[redeboer]

[ ] Add test to verify coupling for parity conserving amplitudes
[ ] Add test to verify decoupling for parity violating amplitudes

Simplest would be to add some new channel test (tests/channel), for instance @sebastianJaeger's channel (versus what?). Better would be a proper unit test where you don't need to run the whole StateTransitionManager, but that is difficult to do.

[redeboer]

Good job!
I'm only thinking about the tests. Currently, only the expected number of parameters is tested. What can happen at most then is that the number changes and you have to figure out whether that change in number makes sense (could come from anything, e.g. #417 would have triggered that). So best would be to somehow compare a channel that does couple the parameters with one that don't, even if by expected number of parameters.
But this argument goes for many of the tests. Many of them don't really test for any logic, but instead just check whether the framework produces the right numbers.