Guiding function for H,F parametrization

[frankschae]

#48 implemented the information filter (H,F parametrization). We still need to add the associated guiding functions in this parametrization and finish the convert functions to convert messages between the H,F and the \nu,P parametrization.

[fmeulen]

Maybe compare also to solve_and_ll! in https://github.com/JuliaDiffusionBayes/GuidedProposals.jl/blob/master/src/log_likelihood.jl

and ∇logρ(i, x, P::HFcSolver{:outofplace}) = F(P, i) - H(P, i)*x
as in https://github.com/JuliaDiffusionBayes/GuidedProposals.jl/blob/master/src/ode_solver_hfc.jl
where H is now called Gamma. So the guiding term becomes like
outer(g) * (F-Gamma*x)