Minor improvements

[lkdvos]

Some small changes and improvements, mostly to remove unused code and make Aqua happy (happier at least).

There is still a large source of ambiguities related to the following:

• environments -> hopefully improved with Refactor InfiniteEnvironments and GradientGrassmann #224 but should be checked afterwards
• LazySum -> having it a subtype of AbstractVector and defining *(::LazySum, ::Any) and related is a problem
• FiniteMPS constructors seem to be a mess

And piracies related to defining methods for type aliases such as MPSTensor

[codecov]

Codecov Report

Attention: Patch coverage is 79.31034% with 6 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/states/quasiparticle_state.jl	62.50%	3 Missing ⚠️
src/states/finitemps.jl	87.50%	1 Missing ⚠️
src/states/ortho.jl	75.00%	1 Missing ⚠️
src/utility/utility.jl	66.66%	1 Missing ⚠️

Files with missing lines	Coverage Δ
src/MPSKit.jl	100.00% <ø> (ø)
...c/algorithms/excitation/quasiparticleexcitation.jl	65.71% <100.00%> (ø)
src/algorithms/grassmann.jl	82.05% <ø> (ø)
src/operators/abstractmpo.jl	39.06% <100.00%> (-0.79%)	⬇️
src/operators/lazysum.jl	75.00% <100.00%> (ø)
src/operators/mpohamiltonian.jl	90.75% <100.00%> (+0.02%)	⬆️
src/states/finitemps.jl	79.49% <87.50%> (+0.16%)	⬆️
src/states/ortho.jl	85.92% <75.00%> (-0.65%)	⬇️
src/utility/utility.jl	75.32% <66.66%> (+4.91%)	⬆️
src/states/quasiparticle_state.jl	87.17% <62.50%> (ø)

[VictorVanthilt]

These look good to me, can you explain the reasoning for adding the whole _not_missing_type thing?

[lkdvos]

It's an unbound type arguments problem: since Vector{Missing} <: Vector{T} where {T<:Union{Missing,A}} would also satisfy the signature, A would then not be defined. Aqua complains about that, even though realistically you would never call this with a vector of missing type. In order to get around this, I need to distill A from the eltype, but it's not easy since you cannot dispatch really dispatch on union types too well.