Initialize NUMBA_DEFAULT_NUM_THREADS with a batch scheduler aware value - #7088

$ cat test_7088.py 
from numba import njit, prange, get_num_threads
import numpy as np

@njit(parallel=True)
def foo(n):
    acc = 0
    for x in prange(n):
        acc += x
    return acc

foo(100)
print(f"get_num_threads(): {get_num_threads()}")

$ python test_7088.py # default, 4 cores on this machine, all accessible
get_num_threads(): 4

$ taskset 0x00000003 python test_7088.py # restrict to cores #0 and #1.
get_num_threads(): 2