Add type annotations

[basnijholt]

I have used MonkeyType when running the tests that automatically added the type hints.

Then I went over all the changes and fixed (or simplified) the things.

[basnijholt]

@akhmerov and @jbweston, I am done with this.

The only thing that has room for improvement, is the loss functions of the LearnerND. However, since we might deprecate that code, I didn't invest too much time.

[codecov-io]

Codecov Report

Merging #244 into master will increase coverage by 0.22%.
The diff coverage is 96.48%.

@@ Coverage Diff @@## master #244 +/- ##
==========================================
+ Coverage 79.52% 79.74% +0.22% 
==========================================
Files 32 32 Lines 4401 4469 +68 Branches 810 810 ==========================================
+ Hits 3500 3564 +64 - Misses 776 778 +2 - Partials 125 127 +2

Impacted Files	Coverage Δ
adaptive/_version.py	48.14% <100%> (+0.48%)	⬆️
adaptive/learner/skopt_learner.py	59.7% <100%> (+1.23%)	⬆️
adaptive/learner/learner1D.py	91.89% <100%> (-0.44%)	⬇️
adaptive/learner/learner2D.py	78.49% <100%> (+0.1%)	⬆️
adaptive/utils.py	83.05% <100%> (+0.59%)	⬆️
adaptive/learner/triangulation.py	85.2% <100%> (+0.31%)	⬆️
adaptive/learner/balancing_learner.py	74.46% <100%> (+0.69%)	⬆️
adaptive/learner/data_saver.py	88.88% <100%> (+0.65%)	⬆️
adaptive/learner/integrator_learner.py	88.17% <100%> (+0.23%)	⬆️
adaptive/learner/integrator_coeffs.py	95.34% <100%> (+0.05%)	⬆️
... and 8 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 dfeb5ef...100f289. Read the comment docs.

[jbweston]

I've put this on my review backlog

[basnijholt]

Closing.

This is superseded by #370, #371, #372, #373, #374, #375, #376, #377, #378, and #379.