Merging models

[SouthEndMusic]

Fixes #912.

The general design is as in the above mentioned issue. 2 things that are kept in mind when merging model B into model A:

• B itself should not be modified, so a copy of it is made and this copy is modified. This modified B is then passed on to merge methods on node and table level;
• Only the merge method on the model level makes modifications, not the ones on node and table level. Only the table merge method does asserts for node ID conflicts (so allocation network IDs are not checked, as these are not unique. Also node/edge geometries are not checked).

[SouthEndMusic]

@evetion mypy still complains about a few things, I don't know what the best way would be to fix them.

[SouthEndMusic]

I'll pick up this model in #851 for testing.

[evetion]

This should ideally not be part of the PR (but will ignore it here)?

[SouthEndMusic]

You mean the whole subnetwork plotting thing? I guess it would have bean cleaner if that was done in a separate PR, but it did help with debugging of the model merging

[SouthEndMusic]

To do:

• Process comments by Maarten above
• Remove Model.merge options: offset_node_id and offset_allocation_network_id are always computed
• offset_spatial: make functions in tests
• inplace: don't make copy of entire model, only loop over nodes and modify those
• concat_models Ribasim-NL#49

[evetion]

• Remove special code to merge initial conditions (default states, after it's removed by another PR)
• Make merge methods on lower levels (node, tables) private
• Make use of model.copy() and check whether it is needed everywhere (don't use a keyword deep=True, just do the right thing)
• Make inplace merge private (only merge (which copies) on model level)

[sonarcloud]

Quality Gate passed

The SonarCloud Quality Gate passed, but some issues were introduced.

2 New issues
0 Security Hotspots
No data about Coverage
0.0% Duplication on New Code

See analysis details on SonarCloud

[visr]

This is paused, after #1110 we can decide how to proceed with this.

[visr]

Closing this PR, since so many things changed that it probably needs a new start. The PR can still be used as inspiration for #912.