When optimizing conditions like 'a = <some_val> OR a IS NULL' so that they're
 united into a single condition on the key and checked together the server must 
 check which value is the NULL value in a correct way : not only using ->is_null 
 but also check if the expression doesn't depend on any tables referenced in the 
 current statement. 
 This additional check must be performed because that optimization takes place 
 before the actual execution of the statement, so if the field was initialized 
 to NULL from a previous statement the optimization would be applied incorrectly.


mysql-test/r/select.result:
  Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
   - test case
mysql-test/t/select.test:
  Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
   - test case. 
     Note that ALTER TABLE is important here : it happens to
     leave the Field instance for t1.b set to NULL, witch is vital for
     demonstrating the problem fixed by this changeset.
sql/sql_select.cc:
  Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
   - check whether a value is null taking into account its table dependency.