Changed order of functions, apparently caused a problem for some
compiler(s?).

Added a toggle button that sets a default filtersize for the
currently selected filter. It is on by default.
Bugfix: RiOption() referenced a blender short value, but RiOption()
expects an integer. This might possibly explain some reported crashes on PPC.

Orco coords on subd.meshes finally work correctly.
They still do not *look* the same as in Blender though,
because QD still only linearly interpolates values.
Changed test of normal/displace flag again, since normals
were still being calculated if the displacement texture
was also used for other purposes. Now bumpnormals are only
calculated if bumpmapping is in fact enabled ('nor' button).
This will still cause severe artifacts in some cases.
If you use displacement, do not enable the same channel
for bumpmapping as well.
Also a fixed a bug that could cause segfaults because
of using a NULL pointer.

Added support for subsurf orco coords, but still doesn't work properly
yet. Some problem in Blender, qdune itself seems to work correctly
with vertex/varying (but still only linear interpolation though).
Multi UV textures now work correctly on subsurf too.

Forgot yet again to undo some testcode...

Support for creases in the subdivsion mesh code has been activated.
While already coded more than a year ago, it was never used.
Fortunately it actually works...
Also added support for semi-sharp creases, which is another quick
inspired hack, but sofar it seems to work reasonably well.
It probably needs some tweaking to make it produce results similar
to Blender though, so it probably will be recoded at some point.
This is not yet usable from Blender, but for the standalone version
of QDune I added another example ribfile, see 'creasetest.rib' in
the ribs/ directory.

Forgot some temporary code.

Preliminary support for 'Arbitrary Output Variables'.
Hopefully this will be useful to implement render passes
and maybe also renderlayers somehow.
This is very early code, and it is not quite complete yet,
but at least it won't break anything (I hope...)
For the standalone version of QD there is an example rib
file 'blsuz_AOV.rib' as a simple demonstration of AOV usage.
The shadercompiler should also now be able to handle shader outputs.
(it already did, but was not handled in the compiled code yet)

Minor modification to avoid valgrind 'uninitialized value' error again.

Minor code modification to still allow the standalone version of QD to
compile without including the blender specific code.