- Examine behavior near 32-bit limit on 32-bit machines.
  (done for factoring)

- segment sieve should itself use a segment for its primes.
  Today we'd need sqrt(2^64) max = 140MB.  Segmenting would yield under 1MB.

- Add test to check maxbits in compiled library vs. Perl

- Figure out documentation solution for PP.pm

- Is the current PP.pm setup the way we want to do it?

- Faster SQUFOF.  We're using the faster racing SQUFOF now, but it could still
  use some tuning.

- Move .c / .h files into separate directory.
  version does it in a painful way.  Something simpler to be had?

- finish test suite for bignum.  Work on making it faster.

- After the factoring changes, we need to use Devel::Cover again to ferret
  out numbers that pass the early tests.

- Test all routines for numbers on word-size boundary, or ranges that cross.

- Test all functions return either native or bigints.  Functions that return
  raw MPU::GMP results will return strings, which isn't right.

- Make proper pminus1 in PP code, like factor.c.

- An assembler version of mulmod for i386 would be _really_ helpful for
  all the non-x86-64 Intel machines.

- Tests for:

   RiemannZeta
   RiemannR

- Dynamically use a mulmodadd in PP aks, just like the new C code does.
  This will mean it'll work for full-size native ints.

- Perl's rand() is a mess.  I believe a decent workaround is to include
  tinymt32, seed it using the system rand (multiple calls, just use 8-bits
  each), then use it to get 32-bit irands.  This would only be used if they
  didn't give us a RNG (so they don't care about strict crypto).

- Add PE 35 Circular primes and PE 291 Panatoipal primes to bin/primes.pl