Optimize validation for large datasets

**Feature request**
It would be great to have analytics for models.

**Use case**
Many users need this functionality for production deployments.

**Implementation idea**
Add a new config option `enable_analytics` with default `false`.

Start by clarifying what analytics for models should measure and how it relates to validation on large datasets. Then locate the configuration and model-validation entry points and define the behavior for enable_analytics when false and true. Done should include a default of false and tests covering the option and its analytics behavior.

Written by the indexing model from the issue text.