Bugfix in vcat.

[GunnarFarneback]

This pull request fixes a concatenation failure.

   _       _ _(_)_     |  A fresh approach to technical computing
  (_)     | (_) (_)    |  Documentation: http://docs.julialang.org
   _ _   _| |_  __ _   |  Type "?help" for help.
  | | | | | | |/ _` |  |
  | | |_| | | | (_| |  |  Version 0.5.0-dev+376 (2015-09-25 02:50 UTC)
 _/ |\__'_|_|_|\__'_|  |  Commit d233ef3* (0 days old master)
|__/                   |  x86_64-linux-gnu

julia> U = UInt(1):UInt(2); [U;[U;]]
ERROR: MethodError: `similar` has no method matching similar(::UnitRange{UInt64}, ::UInt64, ::Tuple{UInt64})
Closest candidates are:
  similar(::Range{T}, ::Type{T}, ::Tuple{Vararg{Integer}})
  similar(::AbstractArray{T,N}, ::Any)
  similar(::AbstractArray{T,N}, ::Any, ::Int64...)
  ...
 in similar at abstractarray.jl:202
 in vcat at abstractarray.jl:745

It is not a regression from 0.3; the same thing happens there.

[ViralBShah]

OS X build failure is the backtrace issue.

[ViralBShah]

Should we backport this to 0.3 as well - assuming there will be one more release?

[tkelman]

Turns out this depends on #13252, which caused an intermittent test failure on master so I won't be backporting either at this time, until that gets sorted out.