[Bf-python] tp_getset update

[Ken Hughes]

I've update the API Update wiki with some things I noticed in reading 
the source code, and plan today to add some to the "methods where we can 
replace string arguments with dictionary constants" part.  I'm kind of 
wading in place until I more general approval on what I've done in 
Lamp.c and Material.c before submitting further patches (I've got 
Camera.c, Image.c, and Lamp.c done).

I've skipped the review of some modules which have outstanding patches 
in the tracker or are known to be very out-of-date; my Curve/IPO ones, 
sapir's MTex and the whole Armature/Bone/NLA stuff.  And for now I'm 
avoiding NMesh.c, hoping someone implements ThinMesh first (ha).

This leaves other ones I'd like to work on but need some discussion 
about the overall goals for the update.  I'll post two here and put the 
rest on the wiki (i.e., this posting is to draw people's attention to 
the wiki and help fill it in).

* I've noted two modules where it would make sense to implement sequence 
or index operators; Metaball.c and Lattice.c.  For example: 
Lattice.getPoint(index), Lattice.setPoint(index,position) and 
Lattice.latSize.  These could be replaced with built-in operators and 
attributes: Lattice.point[i] and len(Lattice.points)
* Any reason for Effect.c and Particle.c to be separate anymore?  I 
think I missed some discussion, but Wave.c is gone and don't recall 
seeing a Build.c ever, so there seems to be just one effect now. 
Doesn't seem to make sense to have a separate BPy Particle object, 
especially the way it's implemented -- as a pseudo derived class.

Ken