Implement sparse operations with UniformScaling using broadcast.

[maleadt]

julia> S = CuSparseMatrixCSR(sprand(2,2,0.5))
2×2 CuSparseMatrixCSR{Float64, Int32} with 2 stored entries:
 0.9425289370752049   ⋅ 
 0.9999923955293779   ⋅ 

julia> I - 3 * S
2×2 CuSparseMatrixCSR{Float64, Int32} with 3 stored entries:
 -1.8275868112256148   ⋅ 
 -2.9999771865881337  1.0

cc @ChrisRackauckas

[codecov]

Codecov Report

Merging #1390 (17709e8) into master (1c75de7) will decrease coverage by 0.88%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1390      +/-   ##
==========================================
- Coverage   77.64%   76.76%   -0.89%     
==========================================
  Files         121      122       +1     
  Lines        8966     8998      +32     
==========================================
- Hits         6962     6907      -55     
- Misses       2004     2091      +87     

Impacted Files	Coverage Δ
lib/cusparse/broadcast.jl	42.15% <100.00%> (-0.35%)	⬇️
lib/cusparse/interfaces.jl	86.41% <100.00%> (+1.48%)	⬆️
lib/cudadrv/occupancy.jl	75.86% <0.00%> (-15.81%)	⬇️
lib/cublas/util.jl	83.67% <0.00%> (-8.17%)	⬇️
src/initialization.jl	62.50% <0.00%> (-7.27%)	⬇️
lib/curand/CURAND.jl	93.75% <0.00%> (-6.25%)	⬇️
src/compiler/exceptions.jl	88.23% <0.00%> (-5.89%)	⬇️
lib/cudadrv/execution.jl	89.65% <0.00%> (-5.18%)	⬇️
lib/cusolver/multigpu.jl	91.17% <0.00%> (-5.15%)	⬇️
lib/cusolver/linalg.jl	84.37% <0.00%> (-4.69%)	⬇️
... and 35 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1c75de7...17709e8. Read the comment docs.