test_utils to make it easier for downstream implementations to check correctness

[st--]

PR type: new feature

Summary

A common use-case of GPflow is to implement your own down-stream kernels / models / likelihoods / etc.
We should make it easy for such users to check their implementation for basic correctness (e.g. kernel is pos-def on test inputs) and adherence to interfaces (accepts tensors of appropriate shapes & returns the right things).

Proposed changes

• Provide a test_utils submodule with basic checks for interfaces.

Minimal working example

import gpflow
from gpflow.test_utils import test_kernel

class MyKernel(gpflow.kernels.Kernel):
    # ...

test_kernel(MyKernel())

Release notes

Fully backwards compatible: yes

PR checklist

• New features: code is well-documented
  • detailed docstrings (API documentation)
  • notebook examples (usage demonstration)
• The bug case / new feature is covered by unit tests
• Code has type annotations
• Build checks
  • I ran the black+isort formatter (make format)
  • I locally tested that the tests pass (make check-all)
• Release management
  • RELEASE.md updated with entry for this change
  • New contributors: I've added myself to CONTRIBUTORS.md

[st--]

Should we provide default arguments for X and X2?

[codecov]

Codecov Report

Merging #1825 (a0274a1) into develop (1a77914) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff            @@
##           develop    #1825   +/-   ##
========================================
  Coverage    98.00%   98.01%           
========================================
  Files          107      108    +1     
  Lines         5761     5785   +24     
========================================
+ Hits          5646     5670   +24     
  Misses         115      115           

Impacted Files	Coverage Δ
gpflow/__init__.py	100.00% <ø> (ø)
gpflow/test_utils.py	100.00% <100.00%> (ø)

---

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 1a77914...a0274a1. Read the comment docs.

[jesnie]

LGTM