[BlockSparseArrays] Redesign nested views

[mtfishman]

To-do:

• Fix tests marked as @test_broken.

Currently, when taking views of views of BlockSparseArrays, the result is sometimes nested views:

julia> using BlockArrays: Block

julia> using NDTensors.BlockSparseArrays: BlockSparseArray

julia> a = BlockSparseArray{Float64}([2, 3], [2, 3])
typeof(axes) = Tuple{BlockArrays.BlockedOneTo{Int64, Vector{Int64}}, BlockArrays.BlockedOneTo{Int64, Vector{Int64}}}

Warning: To temporarily circumvent a bug in printing BlockSparseArrays with mixtures of dual and non-dual axes, the types of the dual axes printed below might not be accurate. The types printed above this message are the correct ones.

2×2-blocked 5×5 BlockSparseArray{Float64, 2, Matrix{Float64}, NDTensors.SparseArrayDOKs.SparseArrayDOK{Matrix{Float64}, 2, NDTensors.BlockSparseArrays.BlockZero{Tuple{BlockArrays.BlockedOneTo{Int64, Vector{Int64}}, BlockArrays.BlockedOneTo{Int64, Vector{Int64}}}}}, Tuple{BlockArrays.BlockedOneTo{Int64, Vector{Int64}}, BlockArrays.BlockedOneTo{Int64, Vector{Int64}}}}:
 0.0  0.0  │  0.0  0.0  0.0
 0.0  0.0  │  0.0  0.0  0.0
 ──────────┼───────────────
 0.0  0.0  │  0.0  0.0  0.0
 0.0  0.0  │  0.0  0.0  0.0
 0.0  0.0  │  0.0  0.0  0.0

julia> @views a[[Block(2), Block(1)], [Block(2), Block(1)]][Block(1, 1)]
3×3 view(view(::BlockSparseArray{Float64, 2, Matrix{Float64}, NDTensors.SparseArrayDOKs.SparseArrayDOK{Matrix{Float64}, 2, NDTensors.BlockSparseArrays.BlockZero{Tuple{BlockArrays.BlockedOneTo{Int64, Vector{Int64}}, BlockArrays.BlockedOneTo{Int64, Vector{Int64}}}}}, Tuple{BlockArrays.BlockedOneTo{Int64, Vector{Int64}}, BlockArrays.BlockedOneTo{Int64, Vector{Int64}}}}, [3, 4, 5, 1, 2], [3, 4, 5, 1, 2]), BlockSlice(Block(1),1:3), BlockSlice(Block(1),1:3)) with eltype Float64:
 0.0  0.0  0.0
 0.0  0.0  0.0
 0.0  0.0  0.0

These nested views can become challenging to deal with. The goal of this PR will be to flatten them down to a single view.

The goal is basically to make use of more of the generic view logic in Base Julia (i.e. in https://github.com/JuliaLang/julia/blob/master/base/subarray.jl), which flattens views using a function reindex. The primary customization point for how reindex behaves is by overloading the to_indices function. BlockArrays.jl has a lot of useful to_indices definitions for Block and other related types in https://github.com/JuliaArrays/BlockArrays.jl/blob/master/src/views.jl, but some are missing for operations we want, so the goal here is to write a minimal number of to_indices overloads to complement the ones already in Base or BlockArrays.

This is a follow-up to #1503 but will provide a more systematic fix to those kinds of issues. Additionally, the goal is to fix a few of slicing bugs listed in ITensor/BlockSparseArrays.jl#2, and also those marked as broken in the BlockSparseArrays tests: https://github.com/ITensor/ITensors.jl/blob/v0.6.14/NDTensors/src/lib/BlockSparseArrays/test/test_basics.jl.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 60.78%. Comparing base (d3afdb7) to head (078105c).

❗ Current head 078105c differs from pull request most recent head 4873cab

Please upload reports for the commit 4873cab to get more accurate results.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

❗ There is a different number of reports uploaded between BASE (d3afdb7) and HEAD (078105c). Click for more details.

HEAD has 3 uploads less than BASE

| Flag | BASE (d3afdb7) | HEAD (078105c) | |------|------|------| ||4|1|

Additional details and impacted files

@@             Coverage Diff             @@
##             main    ITensor/ITensors.jl#1504       +/-   ##
===========================================
- Coverage   78.06%   60.78%   -17.29%     
===========================================
  Files         148      148               
  Lines        9655     9646        -9     
===========================================
- Hits         7537     5863     -1674     
- Misses       2118     3783     +1665     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.