Function safe_arch::load_unaligned_m256

#[must_use]pub fn load_unaligned_m256(a: &[f32; 8]) -> m256

This is supported with target feature avx only.

Load data from memory into a register.

assert_eq!(
  load_unaligned_m256(&[8.0, 17.0, 6.0, 5.0, 1.0, 2.0, 3.0, 4.0]).to_array(),
  [8.0, 17.0, 6.0, 5.0, 1.0, 2.0, 3.0, 4.0]
);