Potts edits

[borisdevos]

generalise the Potts operators for arbitrary Q, as well as tests to check these for Q=4 and Q=5.

[codecov]

Codecov Report

Attention: Patch coverage is 84.31373% with 8 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/operators/spinoperators.jl	89.58%	5 Missing ⚠️
src/models/hamiltonians.jl	0.00%	3 Missing ⚠️

Files with missing lines	Coverage Δ
src/models/hamiltonians.jl	21.81% <0.00%> (+1.81%)	⬆️
src/operators/spinoperators.jl	90.57% <89.58%> (+10.85%)	⬆️

... and 1 file with indirect coverage changes

[lkdvos]

This is definitely a great first PR!
I have just a couple remarks before we can get this merged:

Could you run the formatter, such that the formatter check is happy? This is done file-by-file in VSCode by rightclicking the file and selecting the format option, or for an entire directory (and its subdirectories) by:

using JuliaFormatter
format("path_to_directory")

I think we could consider renaming the operators a bit. U_matrix and V_matrix don't give too much information, and could be anything, so I would suggest:
potts_X and potts_Z, as generalizations to pauli_X etc.

Note that our convention is to have XX + Z as the Ising model, as Z is diagonal in the symmetry basis, so this would mean writing the model as \sum_i potts_XX^i + potts_Z^i or something to keep the consistency. (ie, potts_X should be the non-diagonal one in the symmetric basis!). If this is very strange, we can also consider changing the convention in the ising model to have X be diagonal, but that is somehow also strange to me.

I think it would be very nice if the operators themselves could be tested as well. I am mostly considering Z^q = 1, and the commutation relations, which should be relatively straightforward to test? See also this file

Finally, it is nice for discoverability to have them added to the docs as well. Probably somewhere here would be perfect.