Currently compiles and passes tests on x86_64 with arch_32 and
DECAF_FORCE_32_BIT=1 (as well as the native settigs of course),
so that's a start.

Want to make serialization routine cross-arch.  Need to check that
perf is good enough (likely).  Current routine in p25519/arch_32
is almost cross-arch, but has known bugs (FIXMEs).  Needs to take
into account separate p and, for NEON, the LIMBPERM.

Want to decouple arches for each curve/field.  Currently the split
between decaf_word_t and word_t makes this fraught with peril.  Fix
is probably to rename decaf_word_t to decaf_api_word_t and fix it
to either uint32 or uint64, then make internal things separate per
field.  That way we don't have to try arch detection in the header,
which is nice.

Need to make decaf_gen_tables use SC_LIMB.  Might as well get rid
of API_NS there too.

This is largely a save-your-work checkin.

Created p521/arch_ref64 code to make sure E-521 basically works.
Fixed some of the testing code around E-521.  It doesn't quite pass
everything yet.

Created p521/arch_x86_64 code with optimized multiply.  In this
checkin, the multiply is fast and works, but all the other code in
that directory is the completely unoptimized ref64 build which
reduces after every add and sub.  So the whole thing isn't fast yet.