FIX: Remove force_path_ccw of contours

[greglucas]

This was added to fix previously unoriented paths from Matplotlib's contouring code. This has seemingly been fixed upstream somewhere in the past and there is no test/reproducer in Cartopy, so we can remove this for now because the polygons are oriented as Shapely expects from Matplotlib.

closes #2288

If anyone has ideas for good test cases, it would be nice to add some, but I haven't really given much thought to what produces the bad geometries. An example of what I think would be nice to test below with some minimal values.

1. Contour with a hole in middle of projection
2. Contour with a hole wrapped around the edge of the globe (set central_longitude=180 in the proj_map call)
3. Others...?

import numpy as np
import matplotlib.pyplot as plt
from cartopy import crs as ccrs


proj_data = ccrs.PlateCarree()
proj_map = ccrs.Robinson()

xx, yy = np.meshgrid([-20, 0, 20], [-20, 0, 20])
levels = (10, 50)
zz = np.abs(xx) + np.abs(yy)

ax = plt.figure().add_subplot(projection=proj_map)

cset = ax.contourf(xx, yy, zz, transform=proj_data, levels=levels)
ax.coastlines()

plt.show()

[rcomer]

I don't know how you begin to write a test for this given that we do not have the data from the original issue. I think it makes sense to remove these line as we have no evidence they are still needed.

However, I think there is more going on with #2288: noting that the latitudes in the example run north-south (which I assume is why the contours end up clockwise), if we flip the latitude direction we can still reproduce the problem plot regardless of whether we use main or this branch.

[greglucas]

I think that is a separate issue/bug then having to do with the wrapping of polygons probably... If we flip the indices in that example to be sorted -180 -> 180 (like we do in the transform_first path) then there is no yellow polygon filling everything.

ind = np.argsort(x, axis=1)
x = np.take_along_axis(x, ind, axis=1)
y = np.take_along_axis(y, ind, axis=1)
v = np.take_along_axis(v, ind, axis=1)

[greglucas]

Here is a good issue with links to others all with some geometry getting wrapped around boundaries incorrectly: #2137
which might have the same effect as what is happening here.

[rcomer]

Let's get this in anyway...