Use memmove() in Unicode functions

Where the upstream OpenLDAP code uses AC_MEMCPY(), use memmove()
instead of memcpy() as the copies frequently involve overlapping
memory regions.  Credit to OSS-Fuzz for discovering one instance of
the issue.

https://github.com/krb5/krb5/commit/5c52ab2252953055e64d9b9855ad64b27eda060e
Author: Greg Hudson <ghudson@mit.edu>
Commit: 5c52ab2252953055e64d9b9855ad64b27eda060e
Branch: master
 src/lib/krb5/unicode/ucdata/ucdata.c   |  4 ++--
 src/lib/krb5/unicode/ucdata/ucgendat.c | 20 ++++++++++----------
 src/lib/krb5/unicode/ure/ure.c         |  8 ++++----
 3 files changed, 16 insertions(+), 16 deletions(-)