`@threads` does not work with `JULIA_COPY_STACKS=1`

[chriselrod]

julia> a = zeros(Int, 100_000);

julia> Threads.@threads for i in 1:length(a)
         a[i] = Threads.threadid()
       end

julia> all(isone, a)
true

julia> versioninfo()
Julia Version 1.9.0-DEV.151
Commit b7b46afcf2* (2022-03-07 22:01 UTC)
Platform Info:
  OS: Linux (x86_64-redhat-linux)
  CPU: 28 × Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, skylake-avx512)
  Threads: 28 on 28 virtual cores
Environment:
  JULIA_COPY_STACKS = 1
  JULIA_NUM_THREADS = 28

julia> Threads.nthreads()
28

First noted in: https://discourse.julialang.org/t/possible-threading-bug-on-m1-max-only-arm-build/77779

Behavior is worse on Julia 1.7, where only thread 1 actually writes, but the array is still divided up among all threads:

julia> a = zeros(Int, 100_000);

julia> Threads.@threads for i in 1:length(a)
         a[i] = Threads.threadid()
       end

julia> all(isone, a)
false

julia> all(<=(1), a)
true

julia> sum(==(1), a)/length(a)
0.03572

julia> versioninfo()
Julia Version 1.7.3-pre.0
Commit 2ca8b0cf3d (2022-02-07 18:56 UTC)
Platform Info:
  OS: Linux (x86_64-redhat-linux)
  CPU: Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-12.0.1 (ORCJIT, skylake-avx512)
Environment:
  JULIA_COPY_STACKS = 1
  JULIA_NUM_THREADS = 28

julia> 1/28
0.03571428571428571