mar-antaya/2025_f1_predictions ? reverse-engineered prompt

Build me a Python project that predicts Formula 1 race results for the 2025 season using FastF1 data, past race results, and current qualifying results. I want it to pull the needed data, clean it up, train a Gradient Boosting model on historical races, and then rank the drivers by predicted finishing time for each race. It should also show the model error using MAE so I can see how accurate it is.

Please keep it simple and easy to run, with one script per race if that fits the current setup, and make sure the output is readable, like a clear winner prediction and the rest of the order. If anything needs current FastF1 or scikit learn docs, look them up online.