support BlockRange(B::AbstractBlockArray)

[jishnub]

This parallels CartesianIndices, and provides a convenient way to iterate over the blocks:

julia> B = mortar(fill(rand(1,1),2,2))
2×2-blocked 2×2 BlockMatrix{Float64}:
 0.575617  │  0.575617
 ──────────┼──────────
 0.575617  │  0.575617

julia> BlockRange(B)
BlockRange(Base.OneTo(2), Base.OneTo(2))

julia> BlockRange(B) |> collect
2×2 Matrix{Block{2, Int64}}:
 Block(1, 1)  Block(1, 2)
 Block(2, 1)  Block(2, 2)

[codecov]

Codecov Report

Merging #285 (0cc808e) into master (b03703f) will increase coverage by 91.91%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##           master     #285       +/-   ##
===========================================
+ Coverage    0.00%   91.91%   +91.91%     
===========================================
  Files          16       16               
  Lines        1462     1485       +23     
===========================================
+ Hits            0     1365     +1365     
+ Misses       1462      120     -1342     

Impacted Files	Coverage Δ
src/blocklinalg.jl	94.44% <ø> (+94.44%)	⬆️
src/abstractblockarray.jl	92.30% <100.00%> (+92.30%)	⬆️
src/blockarray.jl	96.29% <100.00%> (+96.29%)	⬆️
src/blockaxis.jl	95.68% <100.00%> (+95.68%)	⬆️
src/blockindices.jl	83.41% <100.00%> (+83.41%)	⬆️
src/show.jl	100.00% <100.00%> (+100.00%)	⬆️

... and 10 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[jishnub]

I've disabled the allocations test, as this is often unreliable while performing tests