resampleing with Integer or Rational unit rate (1) gives Nyquist frequency error

[anowacki]

Trying to resample a signal to its original rate (i.e., with rate = 1 leads to an error in normalize_freq if rate is an Integer or Rational:

julia> using DSP

julia> resample(1:10, 1) # or resample(1:10, 1//1)
ERROR: frequencies must be less than the Nyquist frequency 1.0
Stacktrace:
 [1] error(s::String)
   @ Base ./error.jl:33
 [2] normalize_freq
   @ ~/.julia/packages/DSP/q9iEF/src/Filters/design.jl:241 [inlined]
 [3] #Lowpass#22
   @ ~/.julia/packages/DSP/q9iEF/src/Filters/design.jl:256 [inlined]
 [4] Lowpass
   @ ~/.julia/packages/DSP/q9iEF/src/Filters/design.jl:256 [inlined]
 [5] resample_filter(rate::Int64, rel_bw::Float64, attenuation::Int64)
   @ DSP.Filters ~/.julia/packages/DSP/q9iEF/src/Filters/design.jl:647
 [6] resample_filter
   @ ~/.julia/packages/DSP/q9iEF/src/Filters/design.jl:628 [inlined]
 [7] resample(x::UnitRange{Int64}, rate::Int64)
   @ DSP.Filters ~/.julia/packages/DSP/q9iEF/src/Filters/stream_filt.jl:732
 [8] top-level scope
   @ REPL[90]:1

Inaccuracy of the 'resampled' signal aside, this at least works as expected with a Real argument:

julia> resample(1:10, 1.0)
10-element Vector{Float64}:
  1.0000535025139883
  2.0001070050279766
  3.000160507541965
  4.000214010055953
  5.000267512569941
  6.00032101508393
  7.000374517597918
  8.000428020111906
  9.000481522625895
 10.000535025139882

I wonder whether it is worth checking whether rate == 1 (which is true for 1.0, 1.f0, 1//1, etc.) first in resample, and simply returning the input (or a copy of it) straight away? If so, I'm happy to submit a PR for that.