Generalize concept of removing OpTypes

[lsschmid]

Description

To not only remove Identities but also other Operations I wrote a function that generalizes that concept and allows one to remove arbitrary operation types.

Checklist:

• The pull request only contains commits that are related to it.
• I have added appropriate tests and documentation.
• I have made sure that all CI jobs on GitHub pass.
• The pull request introduces no new warnings and follows the project's style guidelines.

[lsschmid]

For some reason some other tests fail, not sure why.

Also, probably I don't need the functionality any more, so feel free to delete the PR

[burgholzer]

For some reason some other tests fail, not sure why.

Also, probably I don't need the functionality any more, so feel free to delete the PR

Just from a Quick Look: The main reason for the tests failing is that the previous identity removal would also remove (multi-)controlled identities, while your version only removes single-qubit identities without controls.
That should be rather easy to fix though by adding a special case for the identity.

In general, I'd be in favour of getting this merged. It seems like a useful generalisation.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.6%. Comparing base (2782f37) to head (992da1a).
Report is 98 commits behind head on main.

Additional details and impacted files

@@          Coverage Diff          @@
##            main    #624   +/-   ##
=====================================
  Coverage   91.6%   91.6%           
=====================================
  Files        148     148           
  Lines      14731   14736    +5     
  Branches    2365    2365           
=====================================
+ Hits       13499   13504    +5     
  Misses      1232    1232           

Flag	Coverage Δ
cpp	91.3% <100.0%> (+<0.1%)	⬆️
python	99.7% <ø> (ø)

Files with missing lines	Coverage Δ
src/CircuitOptimizer.cpp	88.9% <100.0%> (+<0.1%)	⬆️

... and 1 file with indirect coverage changes

[lsschmid]

Now it should work.
As said, I don't really need it anymore though.

Also: I wrote a comment on the compound operations. Please remove it if it looks fine.

[burgholzer]

Many thanks for the adaptation. I just pushed some small changes. Most notably, switched to unordered_set, which is a hash set and, thus, has fast membership testing (O(1) on average) compared to set (O(log n)). Not that it makes that much of a difference here, but I guess it's still desirable to use the most appropriate data types when easily possible.