Closes #223, #282

This PR adds a boolean temporal and spatial_dim attribute to the CovModel class as well as geo_scale to CovModel and vario_estimate.

Setting temporal=True indicates that the model should be seen as a metric spatio-temporal model. This was possible before by increasing the model dimension by one but this feature basically enables using spatio-temporal model with latlon coordinates. spatial_dim was added so the spatial dimension could be set explicitly and not implicitly by increasing only dim by 1.

Also a geo_scale attribute was added to be able to get meaningful length-scales and bins for latlon models. Before we always used the rescale attribute to do that, but that was rather a hack. Now both can be used to set the sphere radius and to rescale the length-scale. In addition to that I added KM_SCALE, DEGREE_SCALE and RADIAN_SCALE constant that can be used as radius to get a length-scale in km, degree or radian as some people expect the length-scale to be in degree when working latlon coordinates.

Another dimension indicator attribute was added to the CovModel class:

• spatial_dim: The truly spatial dimension of the model (2 for latlon, otherwise dim or dim-1 (if temporal=True))
  • in contrast to dim: the internal model dimension (4 for latlon=True and temporal=True for example)
  • in contrast to field_dim: this is the parametric dimension of the generated field (3 for latlon=True and temporal=True, 2 for latlon=True and temporal=False for example)

Works like this:

import numpy as np
import gstools as gs

model = gs.Matern(
    temporal=True,
    latlon=True,
    len_scale=[1000, 100],
    geo_scale=gs.KM_SCALE,
)

lat = lon = np.linspace(-80, 81, 50)
time = np.linspace(0, 777, 50)
srf = gs.SRF(model, seed=1234)
field = srf.structured((lat, lon, time))
srf.plot()