On the user-defined-authn branch: Remove the character classification
table. It turns out that range checks are faster than table lookups.
Which I should have known.

* src/genctype.py: Removed.
* src/syntax.c:
   - Declare string functions through <apr_want.h>
   - Remove the character classification table and macros.
  (ct_isalnum): New; finds ASCII digits, uppercase and lowercase letters.
  (ct_istoken, ct_istoken68): Reimplement with range checks.
  (skip_space): Likewses; use strspn instead.


git-svn-id: https://svn.apache.org/repos/asf/serf/branches/user-defined-authn@1927498 13f79535-47bb-0310-9956-ffa450edef68