Algorithm for SCF solver including stability analysis

[MarcoEckhoff]

To run an SCF calculation until the stability analysis succeeds, we currently need to configure an entire workflow. Since this workflow follows a standard protocol and is valuable for many applications, it would make sense to provide it as a built-in algorithm. Doing so would significantly enhance the user-friendliness.

A challenging test case for the algorithm is N₂ at an interatomic distance of 4.2 Bohr. The expected total energy is −108.7750566671 Hartree.

[wavefunction91]

@MarcoEckhoff This is an interesting proposal, but how would you handle the underlying modularity of the stability solver in this scheme? This is (intentionally) a customization point.