Broadcasting is broken with Darrays having differing ndims

```julia
julia> d = dzeros((3,3))
3×3 DArray{Float64,2,Array{Float64,2}}:
0.0 0.0 0.0
0.0 0.0 0.0
0.0 0.0 0.0

julia> d2 = dzeros((3,3,1))
3×3×1 DArray{Float64,3,Array{Float64,3}}:
[:, :, 1] =
0.0 0.0 0.0
0.0 0.0 0.0
0.0 0.0 0.0

julia> d .+ d2
ERROR: MethodError: no method matching Base.Broadcast.BroadcastStyle(::Type{Base.Broadcast.DefaultArrayStyle{2}}, ::Type{Base.Broadcast.DefaultArrayStyle{3}})
Closest candidates are:
Base.Broadcast.BroadcastStyle(::Type{T}) where T at broadcast.jl:103
Stacktrace:
[1] Base.Broadcast.BroadcastStyle(::DistributedArrays.DArrayStyle{Base.Broadcast.DefaultArrayStyle{2}}, ::DistributedArrays.DArrayStyle{Base.Broadcast.DefaultArrayStyle{3}}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:21
[2] result_style(::DistributedArrays.DArrayStyle{Base.Broadcast.DefaultArrayStyle{2}}, ::DistributedArrays.DArrayStyle{Base.Broadcast.DefaultArrayStyle{3}}) at ./broadcast.jl:445
[3] combine_styles(::DArray{Float64,2,Array{Float64,2}}, ::DArray{Float64,3,Array{Float64,3}}) at ./broadcast.jl:421
[4] broadcasted(::Function, ::DArray{Float64,2,Array{Float64,2}}, ::DArray{Float64,3,Array{Float64,3}}) at ./broadcast.jl:1263
[5] top-level scope at REPL[79]:1
```

Another one with `reshape`:

```julia
julia> d .+ reshape(d, size(d)..., 1)
ERROR: TaskFailedException:
setindex! not defined for DArray{Float64,3,Array{Float64,3}}
error(::String, ::Type{T} where T) at ./error.jl:42
error_if_canonical_setindex(::IndexCartesian, ::DArray{Float64,3,Array{Float64,3}}, ::Int64, ::Int64, ::Int64) at ./abstractarray.jl:1163
setindex! at ./abstractarray.jl:1152 [inlined]
_setindex! at ./abstractarray.jl:1183 [inlined]
setindex! at ./abstractarray.jl:1153 [inlined]
macro expansion at ./multidimensional.jl:772 [inlined]
macro expansion at ./cartesian.jl:64 [inlined]
macro expansion at ./multidimensional.jl:767 [inlined]
_unsafe_getindex!(::DArray{Float64,3,Array{Float64,3}}, ::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}, ::UnitRange{Int64}, ::UnitRange{Int64}, ::UnitRange{Int64}) at ./multidimensional.jl:762
_unsafe_getindex(::IndexCartesian, ::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}, ::UnitRange{Int64}, ::UnitRange{Int64}, ::UnitRange{Int64}) at ./multidimensional.jl:757
_getindex(::IndexCartesian, ::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}, ::UnitRange{Int64}, ::UnitRange{Int64}, ::Vararg{UnitRange{Int64},N} where N) at ./multidimensional.jl:743
getindex(::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}, ::UnitRange{Int64}, ::UnitRange{Int64}, ::UnitRange{Int64}) at ./abstractarray.jl:1060
(::DistributedArrays.var"#72#74"{Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}},Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3}})(::CartesianIndex{3}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:548
localpart(::DistributedArrays.DestinationSerializer) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/serialize.jl:83
#73 at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:550 [inlined]
construct_localparts(::DistributedArrays.var"#73#75"{DistributedArrays.DestinationSerializer}, ::Tuple{Int64,Int64}, ::Tuple{Int64,Int64,Int64}, ::Array{Int64,3}, ::Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3}, ::Array{Array{Int64,1},1}; T::Nothing, A::Nothing) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:118
construct_localparts at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:118 [inlined]
#137 at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:354 [inlined]
run_work_thunk(::Distributed.var"#137#138"{typeof(DistributedArrays.construct_localparts),Tuple{DistributedArrays.var"#73#75"{DistributedArrays.DestinationSerializer},Tuple{Int64,Int64},Tuple{Int64,Int64,Int64},Array{Int64,3},Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3},Array{Array{Int64,1},1}},Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}}, ::Bool) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/process_messages.jl:79
remotecall_fetch(::Function, ::Distributed.LocalProcess, ::Function, ::Vararg{Any,N} where N; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:379
remotecall_fetch(::Function, ::Distributed.LocalProcess, ::Function, ::Vararg{Any,N} where N) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:379
remotecall_fetch(::Function, ::Int64, ::Function, ::Vararg{Any,N} where N; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:421
remotecall_fetch at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:421 [inlined]
macro expansion at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:88 [inlined]
(::DistributedArrays.var"#1#3"{Tuple{Int64,Int64},DistributedArrays.var"#73#75"{DistributedArrays.DestinationSerializer},Tuple{Int64,Int64,Int64},Array{Int64,3},Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3},Array{Array{Int64,1},1},Array{DataType,1},Int64})() at ./task.jl:356
Stacktrace:
[1] remotecall_fetch(::Function, ::Distributed.LocalProcess, ::Function, ::Vararg{Any,N} where N; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:380
[2] remotecall_fetch(::Function, ::Distributed.LocalProcess, ::Function, ::Vararg{Any,N} where N) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:379
[3] remotecall_fetch(::Function, ::Int64, ::Function, ::Vararg{Any,N} where N; kwargs::Base.Iterators.Pairs{Union{},Union{},Tuple{},NamedTuple{(),Tuple{}}}) at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:421
[4] remotecall_fetch at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.5/Distributed/src/remotecall.jl:421 [inlined]
[5] macro expansion at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:88 [inlined]
[6] (::DistributedArrays.var"#1#3"{Tuple{Int64,Int64},DistributedArrays.var"#73#75"{DistributedArrays.DestinationSerializer},Tuple{Int64,Int64,Int64},Array{Int64,3},Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3},Array{Array{Int64,1},1},Array{DataType,1},Int64})() at ./task.jl:356
Stacktrace:
[1] sync_end(::Channel{Any}) at ./task.jl:314
[2] macro expansion at ./task.jl:333 [inlined]
[3] DArray(::Tuple{Int64,Int64}, ::Function, ::Tuple{Int64,Int64,Int64}, ::Array{Int64,3}, ::Array{Tuple{UnitRange{Int64},UnitRange{Int64},UnitRange{Int64}},3}, ::Array{Array{Int64,1},1}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:83
[4] DArray(::Function, ::Tuple{Int64,Int64,Int64}, ::Array{Int64,1}, ::Array{Int64,1}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:177
[5] distribute(::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}; procs::Array{Int64,1}, dist::Array{Int64,1}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:550
[6] distribute(::Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}) at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/darray.jl:543
[7] _bcdistribute at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:133 [inlined]
[8] bcdistribute at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:129 [inlined]
[9] bcdistribute_args at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:137 [inlined]
[10] bcdistribute_args at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:136 [inlined]
[11] bcdistribute at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:126 [inlined]
[12] copy at /home/jishnu/.julia/packages/DistributedArrays/3FSDt/src/broadcast.jl:93 [inlined]
[13] materialize(::Base.Broadcast.Broadcasted{DistributedArrays.DArrayStyle{Base.Broadcast.DefaultArrayStyle{2}},Nothing,typeof(+),Tuple{DArray{Float64,2,Array{Float64,2}},Base.ReshapedArray{Float64,3,DArray{Float64,2,Array{Float64,2}},Tuple{Base.MultiplicativeInverses.SignedMultiplicativeInverse{Int64}}}}}) at ./broadcast.jl:837
[14] top-level scope at REPL[80]:1
```

Start by reproducing both failures from the examples, then inspect the broadcast paths in src/broadcast.jl and the DArray construction and distribution paths in src/darray.jl named by the stack traces. Done means broadcasting works between Darrays with differing dimensions, including the reshape case, without MethodError or setindex! failures.

Escrito por el modelo de indexación a partir del texto del issue.