Full load hour classes for custom RES potentials

[cpschau]

This PR modifies the preprocessing and overriding process for custom renewable data used in BRIGHT, focusing specifically on onshore wind potentials.

The onshore wind input data have been cleaned so that the full load hours of each data point correlate with its integer step label. Instead of categorizing data points by their steps, the full load hours now determine their assignment to a potential class (onwind, onwind2, onwind3, onwind4). Generally, only two potential classes are used to represent onshore wind potentials (onwind, onwind2), except for the four most favorable regions of the country (BRA_6_1, BRA_17_1, BRA_20_1, BRA_21_1), which are represented by all four classes.

As a result of the new aggregation method, the number of steps serving as the basis for a potential class can vary by location. Currently, the mapping between data points and potential classes follows the quantile distribution:

0 ≤ onwind4 < 0.4 ≤ onwind3 < 0.6 ≤ onwind2 < 0.9 ≤ onwind ≤ 1

The overriding process only occurs at locations with a p_nom_max > 0. Existing capacities retrieved by PowerPlantMatching are set as the p_nom_min for the best potential classes.

Checklist

• I tested my contribution locally and it seems to work fine.
• Code and workflow changes are sufficiently documented.
• Newly introduced dependencies are added to envs/environment.yaml and envs/environment.docs.yaml.
• Changes in configuration options are added in all of config.bright.yaml