Add support for NetCDF/Time-partitioned CSV QLAT inputs

[program--]

This PR adds support for reading NetCDF QLAT outputs from ngen via xarray.

The NetCDF schema for the new outputs are as follows:

netcdf 201512010000NEXOUT {
dimensions:
        feature_id = 3 ;
variables:
        string feature_id(feature_id) ;
                feature_id:description = "Contributing Nexus ID" ;
        double q_lateral(feature_id) ;
                q_lateral:description = "Runoff into channel reach" ;
                q_lateral:units = "m3 s-1" ;
        string time ;
                time:description = "Output timestamp for this contribution" ;
                time:standard_name = "time" ;
                time:format = "%Y%m%s%H%M" ;
data:

 feature_id = "nex-285284", "nex-285285", "nex-285286";

 q_lateral = 8.53977600158463, 7.45141160538848, 5.80901770862798;

 time = "201512010000";
}

The corresponding CSV header is feature_id, <time> (i.e. feature_id, 201512010000).

Related: NOAA-OWP/ngen#605, NOAA-OWP/ngen#612

Additions

• Add NetCDF reading to both AbstractNetwork.read_file and HYFeaturesNetwork.read_file.
• Adds timestamp parsing from file names for *NEXOUT glob filters.

Changes

• Modifies build_qlateral_array when reading a q_lateral file to remove nex- from the prefix.

  This probably should follow the logic of trying to split on -, and remove everything prefixing it...

Testing

• Adds a unit test for reading from ngen-outputted Q Lateral files (both CSV and NetCDF), and ensures forcing sets can be assembled.

Todos

• Add logic for CSV heuristics Unneeded if relying on glob filter
• Add unit tests

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
• Visually tested in supported browsers and devices (see checklist below 👇)
• Project documentation has been updated (including the "Unreleased" section of the CHANGELOG)
• Reviewers requested with the Reviewers tool ➡️

[program--]

@shorvath-noaa if you get a chance, can you pull this PR and ensure the tests and in-memory df schema are as expected -- and maybe do a quick review in case there are things that I missed for this integration? Thanks!!

[shorvath-noaa]

this seems to be entering an infinite loop when I run t-route with *CHRTOUT files as forcing. If you change this to:
qlats_df = nexuses_lateralflows_df.rename(index=self.downstream_flowpath_dict)
it works with *CHRTOUT files and still passes your pytest file.

[program--]

Awesome, I think this works well. Implemented in ebfdd6d.

[program--]

@shorvath-noaa any update on when this can be merged? NOAA-OWP/ngen#612 is soft-blocked on this (but no immediate rush). Thanks!!