Speed up randint array-bounds via chunking Lemire

[vlad-perevezentsev]

This PR implements the word-generation speedup suggested for randint array-like bounds in the #168 (comment)

It changes irk_rand_bounded_broadcast to:

• consume MKL words in cache-sized chunks instead of one full-length pass, avoiding a memory round-trip over the whole array;
• drop the unpredictable lo < s branch in the Lemire loop (t < s always so lo < t already implies lo < s) and memoize the reject threshold per range.

Performance (Intel(R) Xeon(R) Platinum 8480+)

(low=0, high=1e9):

Size (N)	dtype	Before	After	Speedup
5,000,000	int32	52.00 ms	31.44 ms	1.65x
5,000,000	int64	48.82 ms	35.53 ms	1.37x
50,000,000	int32	642.91 ms	428.80 ms	1.50x
50,000,000	int64	602.30 ms	466.51 ms	1.29x