Better support for advanced strategies

[jbaron]

Add more building blocks to make it easier to add advanced strategies like those based on machine learning algo's. Right now requires writing too much plumbing code.

Some required features:

1. Make it easy to capture historic data (multi-dimensional)
2. Add feature engineering capabilities
3. Add bindings to common libraries, like XGBoost

[jbaron]

First steps are now available in a new module called roboquant-ml
Still at prototyping phase to get the API stable.

[jbaron]

Added better feature engineering to the ml module. Right now using Smile for the ML algo's like gradient boosting decision trees.