vectorized linear search for AVX2.c

int vectorizedLinearSearch(unsigned int *array, int size, unsigned int searchNum) {
    assert(((size >> 3) << 3) == size); // divisible by 8 for __m256i
    __m256i searchVector = _mm256_set1_epi32(searchNum); // broadcast searchNum
    for (int i = 0; i < size; i+=8) {
        __m256i searchValue = _mm256_loadu_si256(&array[i]);
        // Would be better if we could compare just a single 32 bit value, but just duplicating
        // the 32 bit value is fine also.
        __m256i eqMask = _mm256_cmpeq_epi32(searchVector, searchValue);
        unsigned int intMask = _mm256_movemask_ps(eqMask);
        if (_mm_popcnt_u32(intMask) != 0) {
            // Get the lowest value found. Necessary to do this because multiple of the same number is fairly common.
            for (int j = 0; j < 8; j++) {
                if (intMask & (1 << j)) {
                    return i+j;
                }
            }
        }
    }
    return -1;
}