Slices and CuArrays

@MasonProtter points out that, somewhat to my surprise, mapslices-like things work on the GPU:

julia> using Zygote, TensorCast, CuArrays, BenchmarkTools
julia> CuArrays.allowscalar(false)
julia> let X = randn(10, 10, 10, 10, 10)
           Xcu = cu(X)
           f(A::AbstractArray{T, 4}) where {T} = reshape(sum(A, dims=(3,4)), size(A,1), size(A, 2))
           g(X) = @cast y[i, j, k] := f(X[:, :, :, :, k])[i, j]
           h_cu(a) = sum(g(a * Xcu))
           h_cpu(a)= sum(g(a * X))
           @show length(X)
           @btime $h_cu'(1)
           @btime $h_cpu'(1)
       end
length(X) = 100000
  1.660 ms (4676 allocations: 200.19 KiB)
  747.911 μs (374 allocations: 3.84 MiB)
351.3479618616162

julia> @pretty @cast y[i, j, k] := f(X[:, :, :, :, k])[i, j]
begin
    local armadillo = sliceview(X, (:, :, :, :, *))
    local porpoise = @__dot__(f(armadillo))
    local jaguar = red_glue(porpoise, (:, :, *))
    y = jaguar
end

It would be nice to (1) make sure this doesn't break, e.g. with #17, (2) understand which cases work or whether some don't, and (3) ideally make it fast?

Start by running the Julia GPU reproduction in the issue and inspect the @cast expansion through sliceview, @dot, and red_glue. Check the cases that work or fail, including compatibility with #17, then verify that supported CuArrays paths remain correct and assess whether the generated operations can be made faster.