See here - roughly speaking, for large plane changes (>38.94 degrees) it is more efficient to burn to a higher apoapsis, do the plane change, then burn down to the original apoapsis again. Said link has all of the math required, including the amount of delta-v required and the optimal apoapsis (for <60 degrees there is a specific apoapsis that is optimal - for >=60 degrees it's optimum to go to infinity, which obviously isn't practical).

Roughly speaking:

1. If you are trying to change your plane by less than cos^-1 (7/9) (again, ~38.94 degrees) , stop, it's not worth it.
2. If you are trying to change your plane by >= 60 degrees, your target apoapsis is "as high as is practical"
3. Otherwise, your target apoapsis is a_c * sqrt(1 - cos theta) / (sqrt(2) - 2 sqrt(1 - cos theta)), where a_c is the semi-major axis of your starting orbit.
4. Burn to said apoapsis at the ascending or descending node between your current orbit and the target orbit
5. Do the plane change at apoapsis
6. Burn at pariapsis to move the apoapsis back down

If I had more time I'd code it myself - I like the modularity of the setup you've got here, it looks relatively easy to do so - but I have far too little time on my hands currently, so I figured I might as well drop this here and see if anything came of it.

I won't have time for this now but as you said it does not look hard. And thanks for providing the math :)