Add tests for C4vCTMRG gradients

[leburgel]

Adds gradient tests for C4v CTMRG contraction.

There seem to be a few combinations that don't work right now. In particular, while the naive gradient_alg = nothing always work, fixed-point differentiation using a :full eigh reverse rule or a diffgauge iteration scheme give incorrect gradients. I marked these cases as broken for now to make sure this is tracked at least. If it's unclear how these can be fixed right now, we can open issues to track this and then merge the tests like this I think.

I also made some minor modifications to the fixed-point differentiation implementations, most notably moving the definition of the iterating function and its pullback outside of the actual leading_boundary_pullback function body. I'm not sure if this makes a difference in practice, but this looks a bit more intuitive to me. I also modified the EigSolver implementation of fpgrad to use VectorInterface.add instead of + and *, to make it easier to use this with other environment types.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/algorithms/ctmrg/gaugefix.jl	97.84% <100.00%> (-0.04%)	⬇️
...rithms/optimization/fixed_point_differentiation.jl	95.50% <100.00%> (+2.32%)	⬆️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

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

[lkdvos]

Looks good to me!

[leburgel]

I opened issues to track the failing cases, so we can merge this to add coverage for the working cases already and then sort out the failures later.