Fixes #1881

Digging into the implementation of proj=cea, it appears that
k_0 and lat_ts are intended to be exclusive ways of specifying the
same concept. EPSG only models the variant using lat_s.
So if k_0 is found and lat_ts is absent, compute the equivalent
value of lat_ts from k_0.

Note: k_0 should normally be in the [0,1] range. In case creative users
would use something outside, we raise an exception, even if the cea
implementation could potentially deal with any k_0 value. Hopefully
this is a (reasonable) limitation that will address nominal use cases.

In the note on the sweep angle, add a line explaining why the sweep angle ultimately affects the geolocation, as it is not immediately or trivially obvious to see why it should matter what the outer or inner gimbal is.  To this end, link to a post on Mathematics Stack Exchange that explains this very well.

Previous version was describing oea by mistake. It is also a projection
by Snyder and the isea docs was based on the paper on the oea
projection. This commit fixes that.

This is a super confusing topic, but from experiments, and in particular with
the example of Fairgrieve projection, o_lat_p and o_lon_p are the coordinates
of the North pole of the unrotated geographic CRS expressed in the rotated
geographic CRS.
This is different from defining a 'new pole', which would be the coordinates
of the rotated North pole expressed in the unrotated CRS.