The idea is to use TABLE_LIST::lock_type for passing type of lock for
target table to mysql_load() instead of using LEX::lock_option 
(which were rewritten by first subselect in SET clause).

This should also fix potential problem with LOAD DATA in SP
(it is important for them to have right lock_type in the table
 list by the end of statement parsing).


mysql-test/r/loaddata.result:
  Added nice test for LOAD DATA with subquery.
mysql-test/t/loaddata.test:
  Added nice test for LOAD DATA with subquery.
sql/log_event.cc:
  Now we don't pass type of lock for target table to mysql_load()
  explicitly . Instead we use TABLE_LIST::lock_type for this table
  which is already properly set here.
sql/mysql_priv.h:
  Now we don't pass type of lock for target table to mysql_load()
  explicitly . Instead we properly set TABLE_LIST::lock_type for
  this table in parser.
sql/sql_load.cc:
  Now we don't pass type of lock for target table to mysql_load()
  explicitly . Instead we properly set TABLE_LIST::lock_type for
  this table in parser.
sql/sql_parse.cc:
  Now we don't pass type of lock for target table to mysql_load()
  explicitly . Instead we properly set TABLE_LIST::lock_type for
  this table in parser.
sql/sql_yacc.yy:
  load_data:
    Let us use TABLE_LIST::lock_type for passing type of lock for target
    table to mysql_load() instead of using LEX::lock_option (which will
    be rewritten by first subselect in SET clause).