This has cost me _hours_ to find. There are many clones of the ws2812
and they have slightly different timing characteristics. Some don't work
so well when you get close to the minimum T1H time: a 1-bit may be
interpreted as a 0-bit.
This commit adds support for the nrf52, which is a Cortex-M4 running at
64MHz.
While implementing this, I discovered that the timings for AVR/16MHz and
Cortex-M0 at 16MHz weren't quite right. They do work, but were slightly
out of spec. So I fixed this as well.