Additional fix for bug #45613: handle failures from my_hash_insert

Testing for presence of stuff in a hash inside the function
that's filling in the hash creates chicken-and-egg type of problems.
This results in test suite failures in mysql-pe in debug mode and 
adds bad initialization dependency in 5.1.
Fixed by removing the debug code.

mysys/hash.c

@@ -334,7 +334,6 @@ my_bool my_hash_insert(HASH *info, const uchar *record)
   size_t idx,halfbuff,hash_nr,first_index;
   uchar *UNINIT_VAR(ptr_to_rec),*UNINIT_VAR(ptr_to_rec2);
   HASH_LINK *data,*empty,*UNINIT_VAR(gpos),*UNINIT_VAR(gpos2),*pos;
-  DBUG_EXECUTE("fail_hash_insert",return(TRUE););
 
   if (HASH_UNIQUE & info->flags)
   {