fn unpack_pair(src: u32x8) -> (u32x8, u32x8)
Unpack 32-bit lanes into 64-bit lanes:
(a0, b0, a1, b1, c0, d0, c1, d1)
into
(a0, 0, b0, 0, c0, 0, d0, 0) (a1, 0, b1, 0, c1, 0, d1, 0)