[Bf-python] Python Ipocurve/CurNurb module changes

[Ken Hughes]

Stephen Swaney wrote:

> I added a section to the wiki on how I see the problem.

Great; thanks.

> Focusing on making the public interface for IPOs and Curves is more
> important than whether the underlying implementation uses derived
> classes.

OK, the code I've been working on for the class stuff is easy to
rip out and use elsewhere.  Since we want to share the same methods
for CurNurbs and Ipocurves, how about I put this code into a file
such as curveutil.c which is callable from either?

(Actually, I already inserted all my iterator/sequence/operator[] code
into Ipocurve.c before I thought of doing this; took about 90 minutes
and executes my test scripts successfully).

> I feel very uncomfortable with the idea of modifying Blender's
> internal structs just to make the bpy implementation 'nicer'.  This is
> more of a Blender3 subject.

The modifications are minor; I added one field to the Ipocurve struct
and reordered variables within.  According to discussion on IRC and
reading Ton's docs on-line, this is safe.  Of course, now moot :-)

Ken