finished changes for solving strict aliasing issue

[Lisandro Dalcin]

Not sure if any of you had the time to look at the small example I sended
to show the strict aliasing issue.

Anyway, I finished the changes needed to solve this issue, all C and
Fortran examples pass. If -Wno-strict-aliasing is removed, GCC will
still complain in many parts, but now we are sure that it will not
generate bad code when using PetscObject and derived structures. BTW,
I commented out all thos XXX_Publish() routines. Is this feature used
in any way? perhaps some external code? If this still have to be
supported, perhaps I can change PetscHeaderCreate() to pass the
publish routine.

Could you create a petsc-dev clone in the hg server in order I can
push this for your to review?

Regards,

PD: Many thanks to Stephan for his 'sed' tip. It was of great help for
this to be finished.


-- 
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594