feat: add benchmark for RNG discard operation

[beckermr]

This PR adds a benchmark for the discard operation so that we can test the effect of removing the fori_loop in another PR.

[codspeed-hq]

CodSpeed Performance Report

Merging #132 will degrade performances by 49.07%

_{Comparing bench-rng (6f15cda) with main (671176c)}

Summary

❌ 1 regressions
✅ 20 untouched benchmarks
🆕 2 new benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
🆕	test_benchmark_rng_discard[compile]	N/A	236.8 ms	N/A
🆕	test_benchmark_rng_discard[run]	N/A	6.9 ms	N/A
❌	test_benchmark_spergel_init[run]	83.6 µs	164.2 µs	-49.07%