pgsql: Go back to using a separate method for doing ILIKE for single

Log Message:
-----------
Go back to using a separate method for doing ILIKE for single byte
character encodings that doesn't involve calling lower(). This should
cure the performance regression in this case complained of by Guillaume
Smet. It still leaves the horrid performance for multi-byte encodings
introduced in 8.2, but there's no obvious solution for that in sight.

Modified Files:
--------------
pgsql/src/backend/utils/adt:
like.c (r1.70 -> r1.71)
(http://developer.postgresql.org/cvsweb.cgi/pgsql/src/backend/utils/adt/like.c?r1=1.70&r2=1.71)
like_match.c (r1.17 -> r1.18)
(http://developer.postgresql.org/cvsweb.cgi/pgsql/src/backend/utils/adt/like_match.c?r1=1.17&r2=1.18)