fix: rand for any float types

[ymtoo]

Errors occur when generating random Red Gaussian and Pink Gaussian noise of any float types.

julia> rand(RedGaussian(1000, convert(Float32, 1.0)))
ERROR: MethodError: no method matching _rand!(::Random._GLOBAL_RNG, ::RedGaussian{Float32}, ::Array{Float64,1})

julia> rand(PinkGaussian(1000, convert(Float32, 1.0)))
ERROR: MethodError: no method matching _rand!(::Random._GLOBAL_RNG, ::PinkGaussian{Float32}, ::Array{Float64,1})

After the changes:

julia> rand(RedGaussian(1000, convert(Float32, 1.0)))
1000-element Array{Float32,1}:
 -0.61762047
  0.026895309
 -0.112934574
  ⋮
 -0.33766067
 -0.964278
 -1.4871243

julia> rand(PinkGaussian(1000, convert(Float32, 1.0)))
1000-element Array{Float32,1}:
 1.1732337
 0.46513683
 0.6875517
 ⋮
 1.5279891
 1.1409322
 1.3683038

[codecov-io]

Codecov Report

Merging #19 (3f3cf76) into master (1c1d8fd) will decrease coverage by 0.03%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master      #19      +/-   ##
==========================================
- Coverage   79.02%   78.99%   -0.04%     
==========================================
  Files          10       10              
  Lines         615      614       -1     
==========================================
- Hits          486      485       -1     
  Misses        129      129              

Impacted Files	Coverage Δ
src/rand.jl	90.90% <100.00%> (ø)
src/dsp.jl	98.02% <0.00%> (-0.02%)	:arrow_down:

---

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 1c1d8fd...3f3cf76. Read the comment docs.

[baggepinnen]

You may find the functions zero(T), one(T) useful in place of converting Float64 0,1 to T. zero(T) also does the right thing for arrays, for algorithms that are generic enough to work on those as well.

[ymtoo]

Yes, it's much cleaner to use zero(T), one(T). Thanks!