Stheno

Breaking Changes

This version contains some re-naming, specifically
- `WrappedGP` -> `AtomicGP`,
- `wrap` -> `atomic`, and
- `CompositeGP` -> `DerivedGP`,
which better reflect what these types / functions represent in the context of a `GPPP`.
It's possible that you've never interacted with them, in which case there's nothing to
worry about.

Lots of code has been moved around in order to better organise everything.

A method of `vcat` has been added to build `BlockData` from `GPPPInput`s. This can make
your code look a bit nicer, so you might want to use it.

Additionally, this package no longer depends explicitly upon Zygote or ZygoteRules.
Instead, all AD rules that are needed use ChainRulesCore directly.

Deprecations mentioned in the 0.7 release have also been dropped.

- Deprecate `approx_posterior` in favour of `posterior`. This is being removed because it has been removed in AbstractGPs in favour of `posterior`. It will be entirely removed in the next breaking release.
- Remove some redundant testing infrastructure and tidy up the file structure slightly.

Enable WrappedGP to work with any AbstractGP, not just the GP type.

AbstractGPs now takes care of everything sparsity-related.
Consequently, Stheno no longer tests anything ELBO-related, and the functionality you get
will depend entirely upon which version of AbstractGPs you're using.

Breaking changes

This version contains a large number of _very_ breaking changes -- some of the basics of the package have completely changed. The over-riding concern has been to make Stheno work well with the AbstractGPs API. Of particular note is the new `GPPP` / `GaussianProcessProbabilisticProgramme` type, which represents a collection of processes and treats them as a single `AbstractGP`.

- The `GaussianProcessProbabilisticProgramme` type has been introduced, and is the new recommended way to work with Stheno.jl. See its docstring, README, and the examples directory for details.
- The `model` macro is gone, and has been replaced with the `gppp` macro. This new macro has quite different functionality, and produces a `GPPP` object.
- Stheno's own kernels have been replaced by ones from `KernelFunctions.jl`. This has removed a lot of code from the repo, but completely changes the function calls required to build GPs. To upgrade, consult KernelFunctions.jl for equivalent kernels.
- Stheno's internal GP type has been replaced with a wrapper type `WrappedGP` for `AbstractGP`s.
- Stheno's internal conditioning and approximate-conditioning functionality has been entirely removed, in favour of using AbstractGPs `posterior` and `approx_posterior` directly on entire `GPPP`s. This much simpler approach generally makes for correspondingly simpler code.

New Features

- Kernels defined using `KernelFunctions.jl` should all work now.
- The gppp macro.
- New `GPPPInput` type for working with `GPPP`s.
- `BlockData` exported for working with multiple processes in a `GPPP` at once.

- Fixed performance bug in reverse-mode gradient computation for the `ELBO`, whereby an `O(N^3)` computation happened in cases where it shouldn't.