Simple update for PEPO (3-site version)

[Yue-Zhengyuan]

This PR adds the 3-site version of simple update for PEPO, which deals with Hamiltonians containing up to next-nearest neighbor terms. The J1-J2 model is used to test the algorithm by comparing the calculated finite-temperature energy (both tr(ρH) and ⟨ρ|H|ρ⟩) with high-temperature expansion results in https://doi.org/10.1103/PhysRevB.67.014416 (as @ogauthe did in #215).

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/algorithms/time_evolution/evoltools.jl	97.82% <ø> (ø)
src/algorithms/time_evolution/simpleupdate.jl	100.00% <100.00%> (ø)
src/algorithms/time_evolution/simpleupdate3site.jl	100.00% <100.00%> (ø)
src/operators/infinitepepo.jl	70.83% <100.00%> (ø)

🚀 New features to boost your workflow:

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

[Yue-Zhengyuan]

Below is the energy at various βs for J2 / J1 = 0.5. exp(-βH) PEPO bond dimension D = 7; CTMRG environment χ = 21 for both 1-layer (tr(ρH)) and 2-layer (⟨ρ|H|ρ⟩) approaches.

[lkdvos]

Overall looks good to me. I see you changed from AbstractTensorMap{<:Any,<:Any,1,N} to AbstractTensorMap{<:Any,<:Any,N,1}, is this just more convenient, or was there any special reason for that?

[Yue-Zhengyuan]

The truncation on the 3-site "cluster" is basically an OBC MPS truncation problem, so I decided to follow MPSKit convention in defining tensors in the cluster. (I was not aware of GenericMPSTensors when writing the 3-site SU before) It doesn't reduce the amount of permutations needed, though.