support Unitful.jl

[Ben Arthur]

would be nice if this worked:

julia> using Unitful, StatsBase

julia> fit(Histogram, (1:1000)u"s", (0:100:1000)u"s")
ERROR: MethodError: no method matching push!(::Histogram{Int64, 1, Tuple{StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}}}, ::Tuple{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}})
Closest candidates are:
  push!(::Any, ::Any, ::Any) at abstractarray.jl:3059
  push!(::Any, ::Any, ::Any, ::Any...) at abstractarray.jl:3060
  push!(::DataStructures.CircularBuffer, ::Any) at /groups/scicompsoft/home/arthurb/.julia/packages/DataStructures/59MD0/src/circular_buffer.jl:83
  ...
Stacktrace:
 [1] push!(A::Histogram{Int64, 1, Tuple{StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}}}, a::Tuple{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}, b::Int64)
   @ Base ./abstractarray.jl:3059
 [2] append!(h::Histogram{Int64, 1, Tuple{StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}}}, vs::Tuple{StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}})
   @ StatsBase /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:334
 [3] #fit#162
   @ /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:353 [inlined]
 [4] #fit#155
   @ /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:298 [inlined]
 [5] fit
   @ /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:298 [inlined]
 [6] #fit#166
   @ /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:407 [inlined]
 [7] fit(::Type{Histogram}, ::StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}}, ::StepRange{Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}, Quantity{Int64, 𝐓, Unitful.FreeUnits{(s,), 𝐓, nothing}}})
   @ StatsBase /groups/scicompsoft/home/arthurb/.julia/packages/StatsBase/XgjIN/src/hist.jl:407
 [8] top-level scope
   @ REPL[4]:1

and gave the same answer as without the units:

julia> fit(Histogram, (1:1000), (0:100:1000))
Histogram{Int64, 1, Tuple{StepRange{Int64, Int64}}}
edges:
  0:100:1000
weights: [99, 100, 100, 100, 100, 100, 100, 100, 100, 100]
closed: left
isdensity: false