Add -mbmi to AVX2 and AVX512 compile flags

Add -mbmi to AVX2 and AVX512 compile flags

The AVX2 and AVX512 flags enable BMI2 but not BMI1, and TZCNT is a
BMI1 instruction. GCC emits the rep bsf encoding that executes as
TZCNT on BMI hardware regardless, but clang gates on the feature bit
and emits plain BSF, which is slower on AMD. Every CPU with AVX2 also
has BMI1, so the flag only affects code already behind AVX2 runtime
detection.

Assisted-By: Claude Opus 4.8 (1M context)