Feature parity with GaussianProcesses.jl

[willtebbutt]

In order to design a front-end that is at least as flexible as GaussianProcesses.jl, and to check that we are at feature parity currently (albeit with more boilerplate), we are going to collect some script that use the JuliaGPs ecosystem to implement stuff that the GaussianProcesses.jl interface allows us to implement.

A good starting point is this folder in the GaussianProcesses.jl repo, which contains example interface usage.

• [ ] Classification
• [x] Mauno Loa Time Series: https://juliagaussianprocesses.github.io/AbstractGPs.jl/dev/examples/1-mauna-loa/
• [ ] Plotting GPs
• [ ] Poisson regression
• [ ] Regression with outliers
• [ ] Regression
• [ ] Sparse Approximations

Please edit the above items with links to gists which do the equivalent using JuliaGPs. Please also feel free to add additional TODO items to this list, provided that they are concretely scoped / you provide an implementation.