Function rsa_cortex_m4::umaal

pub fn umaal(hi: &mut Digit, lo: &mut Digit, m: Digit, n: Digit)

Intention is to replace this with the UMAAL assembly instruction on Cortex-M4.

Operation: (hi, lo) = m*n + hi + lo

This works, because (2^32 - 1)^2 + 2*(2^32 - 1) = 2^64 - 1.