pgsql: Fix usage of char2wchar/wchar2char.

Log Message:
-----------
Fix usage of char2wchar/wchar2char. Changes:
- pg_wchar and wchar_t could have different size, so char2wchar
doesn't call pg_mb2wchar_with_len to prevent out-of-bound
memory bug
- make char2wchar/wchar2char symmetric, now they should not be
called with C-locale because mbstowcs/wcstombs oftenly doesn't
work correct with C-locale.
- Text parser uses pg_mb2wchar_with_len directly in case of
C-locale and multibyte encoding

Per bug report by Hiroshi Inoue <inoue(at)tpf(dot)co(dot)jp> and
following discussion.

Backpatch up to 8.2 when multybyte support was implemented in tsearch.

Tags:
----
REL8_3_STABLE

Modified Files:
--------------
pgsql/src/backend/tsearch:
ts_locale.c (r1.7.2.1 -> r1.7.2.2)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/backend/tsearch/ts_locale.c?r1=1.7.2.1&r2=1.7.2.2)
wparser_def.c (r1.14.2.3 -> r1.14.2.4)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/backend/tsearch/wparser_def.c?r1=1.14.2.3&r2=1.14.2.4)