* passing a non-string secret or non-string hash to any
  CryptContext or handler method will now reliably result
  in a TypeError.

  previously, passing hash=None to many handler identify() and verify()
  methods would return False, while others would raise a TypeError.
  other handler methods would alternately throw ValueError or TypeError
  when passed a value that wasn't unicode or bytes.
  the various CryptContext methods also behaved inconsistently,
  depending on the behavior of the underlying handler.

  all of these behaviors are gone, they should all raise the same TypeError.

* redid many of the from_string() methods to verify the hash type.

* moved secret type & size validation to GenericHandler's encrypt/genhash/verify methods.
  this cheaply made the secret validation global to all hashes, and lets
  _calc_digest() implementations trust that the secret is valid.

* updated the CryptContext and handler unittests to verify the above behavior is adhered to.

* better charset & border-case heuristics for identify command
* fuzzy identify helper added for cisco_type7
* added .summary attr for identify to present more info
* benchmark cmdline interface refined

after some thought, realized the 'nthash' name should implement
the raw hash. since bsd_nthash was very rarely used, it shouldn't
present backwards incompatibility issues at this point to go
ahead and rename it.

* new utils.Base64Engine() provides flexible class for encoding arbitrary base64 charmaps.
  it should also be a bit faster than the old h64 module.
* predefined 'h64' instance has mostly the same methods as the old h64 module
  which it takes the place off - so imports should be unaffected.
  (the only the exception of the xxx_dc_xxx methods, which now use the 'h64big' instance)
* replaced utils._blowfish base64 encoding with custom Base64Engine instance
  to reduce code duplication.
* more through unittests for Base64Engine.