Avoid ind2sub in view with Block{1}

[jishnub]

This avoids the internal function _ind2sub in constructing views of AbstractBlockArrays with one Block{1} index, and instead, indexes into a BlockRange to get the block index. As a consequence, a BlockBoundsError is changed to a BoundsError, but the error message is actually improved.
On master

(BlockArrays) julia> B = BlockArray(zeros(6,6), 1:3, 1:3);

(BlockArrays) julia> B[Block(10)]
ERROR: BlockBoundsError: attempt to access 3×3-blocked 6×6 BlockMatrix{Float64, Matrix{Matrix{Float64}}, Tuple{BlockedUnitRange{RangeCumsum{Int64, UnitRange{Int64}}}, BlockedUnitRange{RangeCumsum{Int64, UnitRange{Int64}}}}} at block index [1,4]

This PR

(BlockArrays) julia> B[Block(10)]
ERROR: BoundsError: attempt to access 3×3 BlockRange{2, Tuple{Base.OneTo{Int64}, Base.OneTo{Int64}}} at index [10]

The main part of the message is that the index [1,4] is changed to [10]. The [1,4] arises from a lack of bounds-checking in _ind2sub, and might be tricky to trace back.

Secondly, this PR introduces a constructor for BlockRange that accepts integer block sizes. This parallels CartesianIndices:

(BlockArrays) julia> collect(CartesianIndices((2,3)))
2×3 Matrix{CartesianIndex{2}}:
 CartesianIndex(1, 1)  CartesianIndex(1, 2)  CartesianIndex(1, 3)
 CartesianIndex(2, 1)  CartesianIndex(2, 2)  CartesianIndex(2, 3)

(BlockArrays) julia> collect(BlockRange((2,3)))
2×3 Matrix{Block{2, Int64}}:
 Block(1, 1)  Block(1, 2)  Block(1, 3)
 Block(2, 1)  Block(2, 2)  Block(2, 3)

[codecov]

Codecov Report

Merging #268 (06d4150) into master (e359943) will increase coverage by 0.11%.
The diff coverage is 96.55%.

@@            Coverage Diff             @@
##           master     #268      +/-   ##
==========================================
+ Coverage   91.78%   91.89%   +0.11%     
==========================================
  Files          16       16              
  Lines        1461     1481      +20     
==========================================
+ Hits         1341     1361      +20     
  Misses        120      120              

Impacted Files	Coverage Δ
src/BlockArrays.jl	66.66% <ø> (ø)
src/blockindices.jl	82.63% <90.90%> (+0.11%)	⬆️
src/abstractblockarray.jl	93.58% <100.00%> (+0.16%)	⬆️
src/blockaxis.jl	96.26% <100.00%> (ø)
src/blocklinalg.jl	94.57% <100.00%> (+0.73%)	⬆️
src/pseudo_blockarray.jl	86.84% <100.00%> (+0.11%)	⬆️

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