Fix is to rewrite the MBR::overlaps() function, to compute the dimension of both
arguments, and the dimension of the intersection; test that all three dimensions are the
same (e.g., all are Polygons).

Add tests for all MBR* functions for various combinations of shapes, lines and points.


mysql-test/include/gis_generic.inc:
  Add tests & checks for bug #24563 and bug #24588 - some GIS functions missing in
  5.1; many GIS functions not tested; Overlaps() function was incorrect when MBR shifted
  only along one axis; Overlaps() needs to take dimension of shape into account.
mysql-test/r/archive_gis.result:
  Update test results.
mysql-test/r/bdb_gis.result:
  Update test results.
mysql-test/r/gis.result:
  Update test results.
mysql-test/r/innodb_gis.result:
  Update test results.
mysql-test/r/ndb_gis.result:
  Update test results.
mysql-test/t/gis.test:
  Add tests & checks for bug #24563 and bug #24588 - some GIS functions missing in
  5.1; many GIS functions not tested; Overlaps() function was incorrect when MBR shifted
  only along one axis; Overlaps() needs to take dimension of shape into account.
sql/spatial.h:
  Add MBR::dimension() (map MBR to integral dimension: point -> 0, line -> 1,
  polygon -> 2, invalid -> -1)
  
  Fix MBR::overlaps() to handle MBRs which are shifted on one dimension only, and to
  take MBR dimension into account.  Also, test both within() and contains() predicates (so
  that overlaps(a, b) == overlaps(b, a)).