Kalman filter log-likelihood

[MichaelClerx]

Fixes #433

[ben18785]

@martinjrobins Interesting to compare this to AR1 model as think they're the same although estimation approach is different (you use the Kalman filter to compute the log-likelihood, I use another approach).

[codecov[bot]]

Codecov Report

Merging #713 into master will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##           master   #713     +/-   ##
=======================================
  Coverage     100%   100%             
=======================================
  Files          68     52     -16     
  Lines        7230   5022   -2208     
=======================================
- Hits         7230   5022   -2208

Impacted Files	Coverage Δ
pints/toy/_repressilator_model.py	100% <0%> (ø)	:arrow_up:
pints/_mcmc/_population.py	100% <0%> (ø)	:arrow_up:
pints/toy/_sir_model.py	100% <0%> (ø)	:arrow_up:
pints/_core.py	100% <0%> (ø)	:arrow_up:
pints/_optimisers/_xnes.py	100% <0%> (ø)	:arrow_up:
pints/_mcmc/_mala.py	100% <0%> (ø)	:arrow_up:
pints/_evaluation.py	100% <0%> (ø)	:arrow_up:
pints/_mcmc/__init__.py	100% <0%> (ø)	:arrow_up:
pints/toy/_goodwin_oscillator_model.py	100% <0%> (ø)	:arrow_up:
pints/toy/_cone.py	100% <0%> (ø)	:arrow_up:
... and 58 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 82471f9...d8ac869. Read the comment docs.

[MichaelClerx]

Fixed the git mess I made here. Please delete local copy and re-check out before continuing work on this

[ben18785]

@martinjrobins this is an oldie but a goodie. Think we might as well try and include it, so long as we have some tests written...