Add covariance estimators

Covariance isn't a distance function, as it can be negative. It however, is often used for similarity search over time-series and should be implemented in SimSIMD.

Start by locating SimSIMD's existing estimator or similarity-search entry points and the tests covering them. Determine the intended covariance inputs, numeric types, and result semantics from the surrounding implementation; done means covariance estimators are implemented and their behavior is covered by tests.

Written by the indexing model from the issue text.