Inaccurate CoL (and CoM) calculations with parts that are offset w.r.t their transform origin (b9 proc wings)

[zephyrspace]

I was initially motivated by a case in RO-RP1, where delta wings had a more fore CoL than swept wings that were basically only the leading edge of the delta wing

Over more testing I found that FAR CoL calculations have a lot of weighting towards the transform origin of the wings. To the point where you could have a completely tailless plane with straight wings way ahead of the center of mass

and still have it be stable.

Unfortunately this isn't just limited to the stock CoL indicator and the VAB far screen, but also extends to the flying scene (where the plane acts according to the displayed CoL/CoM). The blue wing surfaces in these screenshots indicate the rough position of the CoL, the plane with the displaced CoL is extremely nose heavy as FAR and the VAB indicate, while the plane with the usual COL is obviously aerodynamically unstable.

I've also attached the craft files that will reproduce this case (these need the latest RP-1 express install)

Test Case 1.txt
Test Case 2.txt

These test cases are ksp craft files but github doesn't allow me to upload those so they've been renamed