PERF: optimize == and != for GeoSeries (GeometryArray.__eq__)

[jorisvandenbossche]

The (in)equality for GeometryArray is defined here:

geopandas/geopandas/array.py

Lines 1656 to 1667 in 167c061

	# If the operator is not defined for the underlying objects,
	# a TypeError should be raised
	res = [op(a, b) for (a, b) in zip(lvalues, rvalues)]
	res = np.asarray(res, dtype=bool)
	return res
	def __eq__(self, other):
	return self._binop(other, operator.eq)
	def __ne__(self, other):
	return self._binop(other, operator.ne)

So this essentially still uses a python for loop acting on scalar shapely geoometries, instead of one of the vectorized ufuncs from shapely.

Now, if we want to ensure to keep this consistent with shapely's geometry __eq__, there is not actually a direct ufunc equivalent (shapely switched to use equals_exact in 2.0.0, but that was reverted because of ignoring th z dimension, shapely/shapely#1732).
But for a future shapely 2.1.0, there will be a ufunc that is equivalent to the scalar __eq__, i.e. equals_identical (shapely/shapely#1760), and then we should ensure to update our code here to use that ufunc instead of the slow python loop.