Use AbstractGPs

[willtebbutt]

No description provided.

[codecov]

Codecov Report

Merging #51 (c6e516f) into master (babba0a) will decrease coverage by 0.54%.
The diff coverage is 83.07%.

@@            Coverage Diff             @@
##           master      #51      +/-   ##
==========================================
- Coverage   90.12%   89.58%   -0.55%     
==========================================
  Files          20       20              
  Lines        1195     1229      +34     
==========================================
+ Hits         1077     1101      +24     
- Misses        118      128      +10     

Impacted Files	Coverage Δ
src/space_time/rectilinear_grid.jl	87.50% <ø> (ø)
src/util/gaussian.jl	77.41% <50.00%> (-2.59%)	⬇️
src/util/linear_algebra.jl	60.00% <60.00%> (ø)
src/gp/posterior_lti_sde.jl	92.45% <66.66%> (-1.67%)	⬇️
src/gp/lti_sde.jl	90.06% <72.72%> (+0.27%)	⬆️
src/util/zygote_rules.jl	64.67% <80.00%> (-0.24%)	⬇️
src/models/lgssm.jl	96.52% <100.00%> (-0.03%)	⬇️
src/models/linear_gaussian_conditionals.jl	97.65% <100.00%> (ø)
src/models/missings.jl	98.21% <100.00%> (+1.78%)	⬆️
src/space_time/pseudo_point.jl	97.19% <100.00%> (+0.04%)	⬆️
... and 10 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update babba0a...c6e516f. Read the comment docs.

[willtebbutt]

The last bit of this changeover is blocked by JuliaGaussianProcesses/KernelFunctions.jl#208 -- I'll need a patch bump before I can merge this.