Calculate ranger performances (resources usage on flatsat, accuracy ...)

[TanguySoto]

This story is part of epic Evaluate the ranger library for batch training with Random Forest.

• Benchmark ranger on the flatsat with realistic amount of data (PD values + labels, maybe 1000 samples minimum?)
• Calculate the classification metrics of the trained model by using the same dataset used in the paper when evaluating the performance of the Mochi models

[georgeslabreche]

@Evridiki-ase I'm closing this story in favor of splitting it into two stories instead:

• Benchmark ranger on the flatsat with realistic amount of data
• Calculate the classification metrics obtained with ranger's Random Forest model by using the same dataset used in the paper