Add option in raster plot to crop around centroids

[ValentinGebhart]

Changes proposed in this PR:

• Add an optional parameter to geo_im_from_array such that the user can input a (relative) distance above which plot locations that are further away than this distance from any centroids are masked (ie, no values plotted). This option is then available in all function that call geo_im_from_array, i.e., _event_plot, hazard.plot_intensity, hazard.plot_fraction, plot_from_gdf, hazard.plot_rp_intensity, impact.plot_rp_imp
• removed bug about the NaN legend in geo_im_from_array

PR Author Checklist

• Read the Contribution Guide
• Correct target branch selected (if unsure, select develop)
• Descriptive pull request title added
• Source branch up-to-date with target branch
• Documentation updated
• Tests updated
• Tests passing
• No new linter issues
• Changelog updated

PR Reviewer Checklist

• Read the Contribution Guide
• CLIMADA Reviewer Checklist passed
• Tests passing
• No new linter issues

[ValentinGebhart]

Example 1: plot_intensity with floods in France

from climada.util.api_client import Client
client = Client()
properties = {"country_iso3alpha": "FRA", "climate_scenario": "historical"}
hazard = client.get_hazard(hazard_type="river_flood", properties=properties)

Results:

hazard.plot_intensity(event=0)

hazard.plot_intensity(event=0, mask_rel_distance=0.01)

hazard.plot_intensity(event=0, mask_rel_distance=0.001)

Example 2: some return period maps with few centroids

plot_from_gdf(ex_impact_with_zeros, colorbar_name=label, title_subplots=col_label)

plot_from_gdf(ex_impact_with_zeros, colorbar_name=label, title_subplots=col_label, mask_rel_distance=0.1)

plot_from_gdf(ex_impact_with_zeros, colorbar_name=label, title_subplots=col_label, mask_rel_distance=0.01)

[ValentinGebhart]

Default is now not to mask anything, so plots will look as before if user does not specify differently. Up for discussion :) @chahank @spjuhel