Insert a sentinel flowpath downstream of terminal nexuses when partitioning subset hydrofabrics (NGWPC-6553)

[PhilMiller]

Prior to this PR, partitioning a hydrofabric in which there was a nexus that had multiple contributing catchments and no downstream catchment could lead to corrupted run results if those catchments were spread across multiple processes.

Fixes: NOAA-OWP#284

Additions

• SentinelFeature type for the placeholder GeoJSON features that have no geography or geometry associated with them

Changes

• Detect potential victim nexuses in the partitionGenerator input, and add sentinel flowpaths downstream of them
• Elaborate the CMake summary print just a bit

Removals

• Bits of dead code

Testing

1. Output from partitionGenerator on test subset hydrofabric from bug report looks as intended

Screenshots

{
    "partitions":[
        {"id":0,
        "cat-ids":["cat-2395", "cat-2404", "cat-1118", "cat-758", "cat-757", "cat-1257", "cat-1344", "cat-1057", "cat-1480", "cat-1962", "cat-1368", "wb-TERMINAL_SENTINEL-nex-759"],
        "nex-ids":["nex-758", "nex-757", "nex-759"],
        "remote-connections":[
            {"mpi-rank": 1, "nex-id": "nex-759", "cat-id": "cat-1002", "cat-direction": "orig_cat-to-nex"},
            {"mpi-rank": 1, "nex-id": "nex-759", "cat-id": "cat-1143", "cat-direction": "orig_cat-to-nex"},
            {"mpi-rank": 1, "nex-id": "nex-759", "cat-id": "cat-1221", "cat-direction": "orig_cat-to-nex"},
            {"mpi-rank": 1, "nex-id": "nex-759", "cat-id": "cat-1418", "cat-direction": "orig_cat-to-nex"}]}, 
        {"id":1,
        "cat-ids":["cat-1354", "cat-1454", "cat-1591", "cat-1002", "cat-1670", "cat-1001", "cat-1143", "cat-1221", "cat-2018", "cat-1858", "cat-1418"],
        "nex-ids":["nex-1001", "nex-1002", "nex-1143", "nex-1221", "nex-1418", "nex-759"],
        "remote-connections":[
            {"mpi-rank": 0, "nex-id": "nex-759", "cat-id": "wb-TERMINAL_SENTINEL-nex-759", "cat-direction": "nex-to-dest_cat"}]}    ]
}

Checklist

• PR has an informative and human-readable title
• Changes are limited to a single goal (no scope creep)
• Code can be automatically merged (no conflicts)
• Code follows project standards (link if applicable)
• Passes all existing automated tests
• Any change in functionality is tested
• New functions are documented (with a description, list of inputs, and expected output)
• Placeholder code is flagged / future todos are captured in comments
• Project documentation has been updated (including the "Unreleased" section of the CHANGELOG)
• Reviewers requested with the Reviewers tool ➡️

[PhilMiller]

Ok, so reading the ngen code that consumes the partitions, I think it will work fine in the presence of these added sentinels. They won't appear in the geopackage, and so they won't affect the simulation besides as communication placeholders.

[PhilMiller]

Same kind of logic appears in the older GeoJSON reading code - catchments considered are the intersection of what's in the source data and what's specified as the subset from the command line or partition file.

[PhilMiller]

I'm thinking that it may be necessary to add comments geo{package,json}::read and/or their call sites to note this code's reliance on them not checking that every ID named in their arguments is actually present in the data.

[PhilMiller]

Per @zhengtaocui

I tested it with 1, 2, 4, and 6 processors. They all ran without an error. Good job!

[cmaynard-ngwpc]

Looks good. Thank you for excellent descriptions and comments.