Make the generalized Steed programme work

[emsr]

I have the old Fortran code COUL*.FOR and a stab at C++.
I found the paper: The Calculation of Spherical Bessel Functions and Coulomb Functions
by A.R.Barnett

This programme could replace cyl bessel, make Coulomb guts, Whittaker guts, parabolic cylinder funcs, ... and do a good job at just about any pair of 2nd order ODE functions.

It would need the following functors: wronskian, and two kinds of continued fraction partial numers and denoms.

It would be another abstract tool.