426 add bnn

[msmiyels]

BNN Object oriented implementation of the bayesian neural network for binary classification.

Designed for additional cases:

• Multiclass could be added
• Regression could be added

⚠️ Needed to modify the environment setup:

• Local Conda installation
• Docker Poetry installation

Each way needed changes in order to get even the environment running. Also added tensorflow-probability.

Works basically as other ML methods from sklearn using the model.fit() approach.

Done before PR:

• [x] Pytest
• [x] Docs
• [x] Pre-Commit
• [x] Added notebook for basic experiments (own data must be provided)

LICENSING Added THRID-PARTY-LICENCSES since some of the core capabilities rely on MIT licensing.

@nmaarnio is it possible to put the bayesian_neural_network.py under MIT license? It's not mandatory but may simplify the re-use of this particular code in other applications/repos.

OTHERS

• "Quick-fixed" the _extract_values_from_raster function since Windows and Linux handle file paths differently. The former code used rsplit("/", 1) explicitly, which resulted in wrong column names on Windows.