Fix bug for CG in MPI

Removed message window from GUI: messages are now part of the graphic window, in a tile above the button bar. This is a first step in experimenting with a less cluttered UI, as decided during the workshop.

- fixed MED node ordering
- polished 64 bit Cocoa Mac OS X FLTK version
- updated copyright

Added automatic adaptation in advection levelset

order 7->9 seem to be wrong : the way shape functions are
computed is maybe numerically unstable
 

- new GetEnv and GetString functions

Add FullDg formulation
Add Stress and InternalPoint (development)

bindings for GmshMessages and full support for bindings of static functions (eg Msg:getCommRank() == Msg.getCommRank() == Msg():getCommRank())

Propose to automatically create .geo file with a Merge command when
trying to append a scripting command to a non-geo file. Got the idea 
after the discussion with Robert Bou. (This is better than scripting
the Merge command IMHO.)

- more work on SLEPc eigensolver integration

add support for petsc solvers

better gestion of remote messages ("a la getdp")

- moved GUI to FlGui + added experimental QGui to test Qt

- fixed most MSVC and gcc compiler warnings

- converted some remaining tabs to whitespace (that's why the patch is
  so large; please try to remember to indent with 2 spaces and not
  tabs when you commit stuff to CVS)

- merged modified patch from Konstantinos Poulios to allow saving
  physical/partition as element tags in mesh formats that don't
  support them natively