Added winkel tripel projection class

[AgentOxygen]

Rationale

Winkel-Tripel is a popular projection that would be used by multiple users (my group at UT and folks at NCAR). It is also the primary projection for the National Geographic Society.

Implications

I have created a new CRS projection object without modifying any code by utilizing existing class objects and the pyproj backend.

Checklist

1. The new class object can be utilized in similar fashion to the other projection classes.
2. I am unsure how to integrate this into the existing test suite, however I imagine it would undergo the same testing as other projections. I am happy to assist with that integration if I could be pointed in the right direction.

This is my first pull request! I am open to input how on to properly contribute.

[CLAassistant]

All committers have signed the CLA.

[greglucas]

Nice, this looks like a great start.

Regarding the tests: Probably add a test similar to one of the other crs's here containing some transform points and initialization expectations: https://github.com/SciTools/cartopy/tree/main/lib/cartopy/tests/crs

Add the projection to the parameterization list here and add a comparison image.

cartopy/lib/cartopy/tests/mpl/test_mpl_integration.py

Lines 161 to 200 in d57fa13

	@pytest.mark.parametrize('proj', [
	ccrs.Aitoff,
	ccrs.EckertI,
	ccrs.EckertII,
	ccrs.EckertIII,
	ccrs.EckertIV,
	ccrs.EckertV,
	ccrs.EckertVI,
	ccrs.EqualEarth,
	ccrs.Gnomonic,
	ccrs.Hammer,
	pytest.param((ccrs.InterruptedGoodeHomolosine, dict(emphasis='land')),
	id='InterruptedGoodeHomolosine'),
	ccrs.LambertCylindrical,
	pytest.param((ccrs.Mercator, dict(min_latitude=-85, max_latitude=85)),
	id='Mercator'),
	ccrs.Miller,
	ccrs.Mollweide,
	ccrs.NorthPolarStereo,
	ccrs.Orthographic,
	pytest.param((ccrs.OSGB, dict(approx=True)), id='OSGB'),
	ccrs.PlateCarree,
	ccrs.Robinson,
	pytest.param((ccrs.RotatedPole,
	dict(pole_latitude=45, pole_longitude=180)),
	id='RotatedPole'),
	ccrs.Stereographic,
	ccrs.SouthPolarStereo,
	pytest.param((ccrs.TransverseMercator, dict(approx=True)),
	id='TransverseMercator'),
	pytest.param(
	(ccrs.ObliqueMercator, dict(azimuth=0.)), id='ObliqueMercator_default'
	),
	pytest.param(
	(ccrs.ObliqueMercator, dict(azimuth=90., central_latitude=-22)),
	id='ObliqueMercator_rotated',
	),
	])
	@pytest.mark.mpl_image_compare(style='mpl20')
	def test_global_map(proj):

A PR you can look at for updating one of the other projections is #1561

[greglucas]

Do we need to pass in the lat_0 parameter even if 0 is the default, and does that even work, it looks like it is lat_1 in the documentation: https://proj.org/en/9.4/operations/projections/wintri.html

[greglucas]

Do we need to do any of the nan handling here like we are doing in the Robinson projection? i.e. should this subclass Robinson instead if it is similar to that?

[AgentOxygen]

I was able to generate maps containing maps with no problem. Are there other tests for nans that I would need to do?