On many architectures, e.g. 68000, x86, the double registers have higher precision 
than the IEEE standard prescribes. When compiled with flags -O and higher, some double's 
go into registers and therefore have higher precision. In one test case the cost 
information of the best and second-best key were close enough to be influenced by this 
effect, causing a failed test in distribution builds.

Fixed by removing some rows from the table in question so that cost information is not
influenced by decimals beyond standard definition of double.


mysql-test/r/group_by.result:
  Bug#27634: Altered test reslut. The only difference in the results is in the 'rows' column.
mysql-test/t/group_by.test:
  Bug#27634: Altered test case to avoid the corner case where excess precision causes 
  non-minimum costs to appear minimal.