Function safe_arch::add_horizontal_m128d

#[must_use]pub fn add_horizontal_m128d(a: m128d, b: m128d) -> m128d

This is supported with target feature sse3 only.

Add each lane horizontally, pack the outputs as a then b.

let a = m128d::from_array([10.0, 50.0]);
let b = m128d::from_array([100.0, 500.0]);
let c = add_horizontal_m128d(a, b).to_array();
assert_eq!(c, [60.0, 600.0]);