Spatila agg xvec

[masawdah]

Hi team,
This pull request utilized the new functionality of XVEC for spatial aggregation , addressing all the previously mentioned issues related to dependencies and Dask-based computation.

[codecov]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (3ffefe7) 80.98% compared to head (276c3af) 83.25%.
Report is 11 commits behind head on main.

Files	Patch %	Lines
...es_dask/process_implementations/cubes/aggregate.py	92.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #217      +/-   ##
==========================================
+ Coverage   80.98%   83.25%   +2.27%     
==========================================
  Files          29       29              
  Lines        1541     1529      -12     
==========================================
+ Hits         1248     1273      +25     
+ Misses        293      256      -37     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[ValentinaHutter]

Thank you for providing this! Nice to have it in xvec directly! I will have a look at the implementation in the next days and see if we can run this on the polygons over Europe as well! I have also been working on testing the options we have for the large European extent again and will update you on these soon! I think, your implementation in xvec is a good general solution, and I will check it for the use case and get back to you :)

[ValentinaHutter]

Thanks a lot for all your effort for this!

I had a closer look at the aggregate_spatial process for the geometries distributed all over Europe. (You can find information about it here: #124 (comment) ) I was trying to find a general solution for the UC, but it seems like the geometries all over Europe really need to be treated specially - so I ended up with a first temporary implementation in our eodc backend. (Let me know, if you are interested in the reasons behind this - then I can describe it more in detail. Basically, the solution is very backend specific and I do not want to see it as a general solution - but I can share the code with you, or describe what I did. )

I think, it would be great to have your implementation in openeo-processes-dask, as it is much more general and should work for any backend.
The only thing that would be nice to add, before we merge it though, would be a test that also checks some of the values for the geometries, since right now, the check only makes sure that the number of dimensions was reduced.

[clausmichele]

@ValentinaHutter probably you missed one specific bit which is back-end specific? Because aggregate_spatial does not allow to provide a url and load the geometries from it. The specific process load_url should be used instead. Maybe comment here since there is an ongoing discussion about this:
Open-EO/openeo-processes#450