We have an intern working on this currently. It is possible to represent trajectories about Lagrange points since version 10.0, but it requires the trajectory to be computed by a numerical model. This opens a can of worms because numerical models clearly show the unstable nature of these trajectories. Halo orbits are cool but really happen only in models where physics have been simplified (the so call CR3BP for Circular Restricted 3-body problem). Reality is quite far from CR3BP.
Our intern has already implemented the CR3BP part, so by now we can find halo orbits in the simplified physics model, and we still can model full trajectories with all perturbations for realistic simulations. His current work is to build a bridge between the two models. Our current idea is to progressively change the simplified model into the full model (think homotopic methods). This work is ongoing.
If you want to have a look at the current state of the project, you need to checkout the specific git repository that has been set up for the intern and look at the cr3bp branch: https://gitlab.orekit.org/vmouraux/orekit/tree/cr3bp.