[Bf-taskforce25] mesh DNA

[joe]

The bmesh branch is now reaching the point where it'll soon be time to
redo mesh DNA.  Currently, I'm planning on adding several new structs
for storing ngon face data; the original MFace structure will be used
to store tessellations, for compatibility reasons.  Once again, I'd
like to remind everyone not to make any major modifications to the
mesh infrastructure in 2.5, since that's a good way to kill my
efforts.

I'm not sure anyone cares about the fine details of what the new
structs will look like; I'll probably take a look at the mesh DNA in
the old bmesh branch and see if I can use it.  It should be fairly
easy to design the structs themselves.

Joe