No test case as the bug is in an existing test case (rpl_trigger.test
when it is run under valgrind).
The warning was caused by memory corruption in replication slave: thd->db
was pointing at a stack address that was previously used by 
sp_head::execute()::old_db. This happened because mysql_change_db
behaved differently in replication slave and did not make a copy of the 
argument to assign to thd->db. 
The solution is to always free the old value of thd->db and allocate a new
copy, regardless whether we're running in a replication slave or not.


sql/log_event.cc:
  Move rewrite_db to log_event.cc, the only place where it is used.
sql/slave.cc:
  Move rewrite_db to log_event.cc
sql/slave.h:
  Remove an unneeded declaration.
sql/sql_class.h:
  Fix set_db to always free the old db, even if the argument is NULL.
  Add a comment.
sql/sql_db.cc:
  Always make a deep copy of the argument in mysql_change_db, even 
  if running in a replication slave. This is necessary because 
  sp_use_new_db (stored procedures) assumes that mysql_change_db always makes
  a deep copy of the argument, and thus passes a pointer to stack into it.
  This assumption was true for all cases except the replication slave thread.