Toroid core Al numbers are determined with uniform distributed winding, similar to the first photo. The low permeability core involves a relative large fringe field for a non-uniform winding. I see no other option than calculating the actual field geometry with a 3D magnetic field solver and determining inductance as an integral value.
I'm not aware of a specialized tool for toroid inductors with non-uniform winding.
- - - Updated - - -
FastHenry is free and my preferred tool for arbitrary conductor configurations, but unfortunately not supporting magnetic cores.