Rationale document for bit-precise types _BitInt

This type has been added into the C2x specification, alongside changes
to describe how they are represented at a machine level we also add a
design document describing the rationale behind the choices we made.

One of the main contentious issues has been around the alignment of
large _BitInt types.  In this document we have chosen to specify that
large _BitInt's are treated as if they were an array of double-register
sized chunks.

Other psABI's have chosen to represent large _BitInt's as an array of
single-register sized chunks.  This means that differences in alignment
might be surprising to developers attempting to write portable code.
However in contrast we believe that developers familiar with ABI's for
Arm Architectures would be less surprised by this decision.

N.b. I happened to also notice a stack overflow question which was
suggesting the use of _BitInt(128) for u128.  Given the discussion did
not have any mention of the different ABI between this and __uint128 I
would take it as evidence for a benefit to having the two integrals ABI
match.
https://stackoverflow.com/questions/16088282/is-there-a-128-bit-integer-in-gcc