Fix two small but very nasty bugs in the PCH writer for method pools:

  (1) Make sure to pad on-disk hash tables with 4 bytes, not 2, since
  the reader assumes that bucket data is aligned on 4-byte
  boundaries. 
  (2) Don't emit the number of factory methods twice. This was
  throwing off the data counts and therefore causing lookups to
  fail. I've added asserts so that this class of error cannot happen
  again.

llvm-svn: 69991