[PeterNelemans]

Issue addressed

Fixes #768 and #767

Explanation

I added another example notebook for building a Wflow model using Python, and improved the one for deriving some data from the DEM.

New: build_model_python.ipynb
The existing build example builds a complete model with a single CLI command. That works, but if something goes wrong it is hard to tell what caused it, and users cannot inspect intermediate results. The new notebook builds a model step by step with the Python API, plotting or inspecting the result after each step:

• explore the data catalog and delineate the subbasin from an outlet point (get_basin_geometry)
• model settings (setup_config), basemaps (MERIT Hydro), rivers (hydro_rivers_lin) and river roughness
• reservoirs with and without control (hydro_reservoirs, hydro_lakes) and glaciers (rgi)
• land use (CORINE), leaf area index (MODIS) and soil (SoilGrids)
• forcing: precipitation (CHIRPS) and temperature and PET (ERA5, De Bruin)
• outlets and gauges from test_stations.csv, with subcatchments
• constant parameters, and finally writing the model with model.write()

Renamed: build_model.ipynb → build_model_cli.ipynb

• Renamed so the two build examples are clearly paired (CLI vs. Python).
• Updated the title and intro, and added a link to the Python example.
• Fixed small typos and grammar.

Reworked: prepare_ldd.ipynb
On main, this notebook takes flow directions from MERIT Hydro, derives a river network from them and uses that to re-derive flow directions, which is a bit circular and not very helpful for most users. It now starts from a DEM only (GEBCO), which is the realistic case:

• derive D8 flow directions from the DEM with hydromt.gis.flw.d8_from_dem
• derive upstream area, Strahler stream order, slope and basins
• compare two river definitions (Strahler order >= 3 vs. upstream area > 10 km²) and explain why there is no one-size-fits-all choice
• export the derived data to examples/gebco_derived_hydrography/ and load it again through a data catalog

Output folders

• The CLI and Python notebooks now write to wflow_test_cli_full and wflow_test_python_full, so they don't overwrite each other when the docs execute them. Updated .gitignore and the build-example-model-full pixi task to match.

Docs

• example_index.rst: renamed the dropdown to "Build model", linked both build notebooks and added build_model_python to the toctree.
• user_guide/2_sbm_model/2_build.rst: updated the references to build_model_cli and added a link to build_model_python.

Checklist

• Updated tests or added new tests (not applicable, documentation only)
• Branch is up to date with main
• Tests & pre-commit hooks pass
• Updated documentation if needed

Additional Notes (optional)

• The notebooks run on artifact_data only, so no extra data is needed.
• The example label example-build_model is now example-build_model_cli. All internal references are updated.

Let me know what you think!