Make C4v rotation and reflection symmetry checks into warnings

[pbrehmer]

Here we swap out ArgumentErrors for @warns in the input checks for C4vCTMRG in case the PEPS is not rotation or Hermitian reflection symmetric. This may be relevant in cases such as #380, where the previous checks were too restrictive.

[codecov]

Codecov Report

❌ Patch coverage is 80.00000% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...rithms/optimization/fixed_point_differentiation.jl	0.00%	2 Missing ⚠️

Files with missing lines	Coverage Δ
src/algorithms/ctmrg/c4v.jl	83.03% <100.00%> (+0.96%)	⬆️
...rithms/optimization/fixed_point_differentiation.jl	89.34% <0.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[lkdvos]

Looks good to me, although I guess the checks will still error for that particular case right?

[pbrehmer]

Looks good to me, although I guess the checks will still error for that particular case right?

In the case of #380, there will still be errors because there I explicitly call the symmetrize! function. I guess there the trick would be to implement a custom symmetrization scheme that won't lead to an error (I guess by flipping the physical space or something).

[pbrehmer]

Well nevermind, indeed it is still erroring. I forgot that _isapprox_localsandwich of course adds the original and symmetrized PEPS tensor and this will make it error due to differing spaces. I guess then I should rather catch that error first before raising a warning.