Feature/span cbar slot based

[cvanelteren]

Continuation of #681
Should close#678 properly

[codecov]

Codecov Report

❌ Patch coverage is 87.45645% with 36 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
ultraplot/axes/geo.py	73.80%	11 Missing and 11 partials ⚠️
ultraplot/axes/base.py	68.18%	8 Missing and 6 partials ⚠️

📢 Thoughts on this report? Let us know!

[gepcel]

Doesn't seem to get the right height of GeoAxes. Try the following code:

import ultraplot as uplt
import numpy as np
fig, axs = uplt.subplots(nrows=2, ncols=2, proj=["npstere", "npstere", 'cyl','cyl'])
# fig, axs = uplt.subplots(nrows=2, ncols=2)
rng = np.random.default_rng()
data = rng.random((100, 100))
cm = axs[0, 0].pcolormesh(data)
fig.colorbar(cm, loc="r", ax=axs[0], ref=axs[1,0])
fig.colorbar(cm, loc="r", ax=axs[0], ref=axs[:,1])

[cvanelteren]

You mean this is not looking correct:

Are you referring to how the bottom ones are reshaping to a different aspect?

[gepcel]

As ot mark ① has the same height of axs[1,0] when colorbar doesn't has span. but ② doesn't has the same height, with span.

[cvanelteren]

I think this works. With the latest commit:

["npstere", "cyl", "cyl", "cyl"]: would adjust but
["npstere", "cyl", "cyl", None]: will not as the last line would have a plot that is longer.

[cvanelteren]

Worst case we have this which I am ok with:

[cvanelteren]

Will let this ruminate a bit to see if I can improve this -- it is a bit janky imo.