Julia 1.9 adaptions

[carstenbauer]

• openblas_get_affinity-based functions, e.g. openblas_getcpuids() (https://github.com/carstenbauer/ThreadPinning.jl/issues/6)
• threadpool filter in threadinfo()

TODOs

• [x] check if @tspawnat implementation is still good

Closes #6

[carstenbauer]

julia> using ThreadPinning

julia> using LinearAlgebra

julia> BLAS.set_num_threads(5)

julia> ThreadPinning.openblas_print_affinity_masks()
|11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111|
|11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111|
|11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111|
|11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111|
|11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111|

julia> ThreadPinning.openblas_pinthreads(:compact)

julia> ThreadPinning.openblas_print_affinity_masks()
|10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000|
|01000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000|
|00100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000|
|00010000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000|
|00001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000|

julia> ThreadPinning.openblas_getcpuids()
5-element Vector{Int64}:
 0
 1
 2
 3
 4

julia> ThreadPinning.openblas_pinthreads(:spread; places=:numa)

julia> ThreadPinning.openblas_getcpuids()
5-element Vector{Int64}:
  0
 16
 32
 48
  1