Merge fkpl collisions

[mrhardman]

Pull request containing the weak-form Fokker-Planck self-collision operator for ions and associated features (check-in tests, underlying ability to support cross-ion-species collisions). Creating this request now to gather feedback before the merge. The major additions are the fokker_planck*.jl modules and the gauss_legendre.jl module, which provides the weak-form matrices necessary for the implementation.

To do list (before merging):

• (COMPLETE) Include gauss_legendre.jl methods in calculus_tests.jl
• (COMPLETE) Tests for second derivative function provided by the gauss_legendre.jl module.
• (COMPLETE) Remove Q from second_derivative functions for chebyshev methods.
• Retest the manufactured solutions tests (@johnomotani's Krook operator appears to break the MMS tests, again).
• (COMPLETE) Clean up the earlier "strong-form" implementation of the collision operator.
• (COMPLETE) Test the collision operator in a domain with sheath boundary conditions to check for reasonable behaviour.
• (COMPLETE) Structure calculus.jl and the chebyshev.jl, gauss_legendre.jl modules in a fashion that is compatible with the new structure in master.
• (COMPLETE) Example input file.
• (COMPLETE) Instructions/checks for when to use "chebyshev_pseudospectral" vs. "gausslegendre_pseudospectral" options.
• (COMPLETE) Determine how to usefully store interactive tests scripts "2D_FEM_assembly_test.jl" "fkpl_functional_test.jl" and "GaussLobattoLegendre_test.jl".
• Any other business (keep this list updated).

To do list (after merging):

• In-code documentation for the collision operator and weak methods. See issue https://github.com/mabarnes/moment_kinetics/issues/152.
• In-code documentation of the collision operator tests and expected behaviour. See issue https://github.com/mabarnes/moment_kinetics/issues/152.
• Unify the functions used to impose boundary conditions on F and the collision operator C[F,F]. See issue https://github.com/mabarnes/moment_kinetics/issues/153.
• Finish the optimised implementation of the MPI (see https://github.com/mabarnes/moment_kinetics/issues/140).
• Support the vperp BC enforcing regularity at vperp =0 for the chebyshev grid. https://github.com/mabarnes/moment_kinetics/issues/157. This issue likely prevents the use of numerical dissipation in the vperp coordinate with chebyshev points.
• Support a nu_ii computed from reference parameters. https://github.com/mabarnes/moment_kinetics/issues/163.

[mrhardman]

I think we should merge this now. 🎉

Let's go for it.

[mrhardman]

Once the tests are run on master and I have the simulation I want we can delete the development branch.