[paper 1] add plot with the difference in return level (and relative difference) for the big shapes. we conclude that there is big difference indeed

[paper 1] add upper and lower bound into the gev param object. Add script to run fit wihout maximum, and then compare the upper bound with the maximum values that was removed. The test is ok for all altitudes, the upper bound found was always higher than the maximum values

[PAPER 1] taking into account psnow when computing TDRL and also when computing return level uncertainties

[EXTREME ESTIMATOR] add test confidence interval. Add confidence interval with the ci method in the Bayesian case. add test for the stationary case. put the division by 100 (to get snow load) in eurocode vizualizer

[EXTREME ESTIMATOR] refactor margin_fits folder. extreme estimator folder. and extreme_models folder

[TREND ANALYSIS] add boolean that accept scale 0 parameter that enables to compute easily some mean difference where scale1 is zero

[PAPER1] modify the strength criterion for the hypercube (use the 0.98 quantile and its evolution in 10 years)

[GEV] add computation of the mean & variance. modify gev_estimator. add function visualize_summary_of_annual_values_and_stationary_gev_fit: to quickly visualize the repartition of the main empirical indicators and corresponding stationary gev indicators.

[EXTREME PARAMETERS] create an attribute undefined_parameters in ExtremeParameters object, to avoid crash when a shape is < 0 for point while we want to display

[EXTREME ESTIMATOR] add margin_fits folder. add gpd params & gpdmle_fit. refactor code to handle gev and gpd together