Add other uncertainty metrics than spatial correlation of errors

[rhugonnet]

Most space agency use classic metrics of spatial error (even though those don't tell so much for DEMs sometimes because of long-range correlations), such as LE90, CE90, and/or SE90 corresponding to 90% probable vertical, horizontal, and 3d spherical radial error. See for instance PGC or TanDEM-X products.

We should include those. Maybe a reference document to use for this would be nice.

Some references we could point towards:
https://iopscience.iop.org/article/10.1088/1681-7575/ab1705#metab1705s3
https://www.asprs.org/wp-content/uploads/Lidar_Guidelines_3-19-2018.pdf
https://pythonhosted.org/uncertainties/
https://www.asprs.org/revisions-to-the-asprs-positional-accuracy-standards-for-geospatial-data-2023