Fix derivatives at the edges of finite operators

[lkdvos]

This is a bugfix for behavior described in #295, which still needs some extra tests.
In particular, we should probably test somewhere that time-evolving a groundstate yields the desired results, which is how I debugged this (as originally shown by #295 as well).

To do

• Add test for phase of evolution of eigenstate
• Add test for handling real inputs

[codecov]

Codecov Report

Attention: Patch coverage is 94.28571% with 2 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/algorithms/timestep/tdvp.jl	80.00%	2 Missing ⚠️

Files with missing lines	Coverage Δ
.../algorithms/derivatives/hamiltonian_derivatives.jl	92.77% <100.00%> (+1.52%)	⬆️
src/operators/jordanmpotensor.jl	81.40% <100.00%> (+0.09%)	⬆️
src/states/finitemps.jl	82.43% <100.00%> (+2.16%)	⬆️
src/algorithms/timestep/tdvp.jl	87.20% <80.00%> (-1.11%)	⬇️

... and 8 files with indirect coverage changes

🚀 New features to boost your workflow:

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

[ZongYongyue]

Currently timestep asks the eltype of H is ComplexF64, otherwise it will throw ERROR: InexactError

[lkdvos]

Is this with the latest version of this branch already? Could you tell me what you are running and what the error message is? I can't seem to reproduce it for now

[edit] found it, should be fixed soon

[ZongYongyue]

Is this with the latest version of this branch already? Could you tell me what you are running and what the error message is? I can't seem to reproduce it for now

[edit] found it, should be fixed soon

It is fine now😄

[VictorVanthilt]

The BlockTensorKit fix is in v0.1.9 and the version in MPSKit's Project.toml is 0.1.6 but somehow the problems have been fixed here?

[lkdvos]

Given that these fixes in blocktensorkit aren't too important, I don't think it's necessary to force everyone to update. The test will automatically run on the latest version so that should do the trick

[VictorVanthilt]

Wow, didn't know the tests run on different versions than the [compat] in the Project.toml

[lkdvos]

I agree that this is not necessarily a great design choice but oh well 🤷