Transport improvements from 2025-W15

[khaeru]

#300 ← previous PR | following PR → #349

• Update and fix the automated workflow for running MESSAGEix-Transport.
• Address KeyError: 'OTHER' in the report step.
• Simplify and improve preparation of dynamic constraint data:
  • Drop individual code for .freight, .ldv, and .passenger submodules.
  • New file constraint-dynamic.csv to unify these values.
  • Add/update values for dynamic constraints on freight technologies.
• Replace the "non-ldv" label with "P[assenger] ex[clusive of] LDV". The former was ambiguous, as it referred to non-LDV passenger transport only, but could be confused as applying to freight (which is also not passenger LDV).
• Use simpler, more clear names for some technologies:
  • FR_FCg → f road gas fc
  • FR_FCm → f road methanol
  • FR_ICG → f road gas ic
• Remove old test XFAIL marks added in Check out correct ref in "pytest" CI workflow #288, Work around khaeru/genno#171 #337, and Cache message-static-data in "pytest" workflow #339; these tests now pass again.
• Adapt to Fix historic vintages in map_tec_lifetime message_ix#924 using the .migrate.v311() utility from Use Zenodo snapshots for nightly scenario tests message_ix#936.
• Update MESSAGEix-GLOBIOM base models to e.g.
  ixmp://ixmp-dev/SSP_SSP1_v5.0/baseline_DEFAULT_step_13
  (from …v2.1…).
• Copy emission price trajectories to MESSAGEix-Transport scenarios #306 Next PR.
• Add a t (freight mode) dimension to the elasticity_f input data flow; adjust freight activity calculation to incorporate these values. Next PR.

Not specific to transport:

• New genno operator select_allow_empty().
• Improve .util.MappingAdapter with an on_missing parameter/option and from_dicts() class method.
• Improve log output from .check.ContainsDataForParameters.
• .tools.iea.web.IEA_WEB: Drop the special handling for node=R12_MEA with transform=B option, added in [TODO link the PR]. We determined that this produces an incorrect aggregation, so it was likely based on incomplete information about the private code referenced there.

How to review

Note that the CI checks all pass.

PR checklist

• Continuous integration checks all ✅
• Add or expand tests; coverage checks both ✅
• Add, expand, or update documentation.
• Update doc/whatsnew.

[codecov]

Codecov Report

Attention: Patch coverage is 89.52703% with 31 lines in your changes missing coverage. Please review.

Project coverage is 77.8%. Comparing base (fdd4e60) to head (cef8c0e).
Report is 51 commits behind head on main.

Files with missing lines	Patch %	Lines
...sage_ix_models/tests/model/transport/test_build.py	65.8%	14 Missing ⚠️
message_ix_models/testing/check.py	0.0%	7 Missing ⚠️
message_ix_models/util/common.py	86.7%	7 Missing ⚠️
message_ix_models/util/importlib.py	92.8%	2 Missing ⚠️
...essage_ix_models/tests/model/transport/test_ldv.py	75.0%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##            main    #328     +/-   ##
=======================================
- Coverage   78.5%   77.8%   -0.8%     
=======================================
  Files        217     219      +2     
  Lines      17432   17495     +63     
=======================================
- Hits       13701   13614     -87     
- Misses      3731    3881    +150     

Files with missing lines	Coverage Δ
message_ix_models/model/transport/base.py	97.9% <ø> (ø)
message_ix_models/model/transport/build.py	95.1% <100.0%> (+<0.1%)	⬆️
message_ix_models/model/transport/config.py	98.6% <100.0%> (-0.1%)	⬇️
message_ix_models/model/transport/constraint.py	100.0% <100.0%> (ø)
message_ix_models/model/transport/data.py	97.5% <100.0%> (+<0.1%)	⬆️
message_ix_models/model/transport/disutility.py	100.0% <100.0%> (ø)
message_ix_models/model/transport/freight.py	100.0% <100.0%> (+17.5%)	⬆️
message_ix_models/model/transport/ldv.py	97.4% <100.0%> (-0.4%)	⬇️
message_ix_models/model/transport/operator.py	85.0% <100.0%> (ø)
message_ix_models/model/transport/other.py	100.0% <100.0%> (ø)
... and 22 more

... and 8 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[khaeru]

I am seeing test failures like:

FAILED message_ix_models/tests/model/transport/test_report.py::test_simulated[True] - ValueError: Couldn't identify node codelist from ['World', 'chicago', 'new-york', 'san-diego', 'seattle', 'topeka']
FAILED message_ix_models/tests/model/transport/test_report.py::test_simulated_iamc - ValueError: Couldn't identify node codelist from ['World', 'chicago', 'new-york', 'san-diego', 'seattle', 'topeka']
FAILED message_ix_models/tests/model/transport/test_report.py::test_simulated_plot[stock-ldv] - ValueError: Couldn't identify node codelist from ['World', 'chicago', 'new-york', 'san-diego', 'seattle', 'topeka']

I think this indicates that the fixtures for these tests, or some other, unrelated tests, are 'leaking' such that make_dantzig() scenario contents end up in the test scenarios needed for these tests. Will debug.

[khaeru]

@r-aneeque if you can leave the ceremonial ✅, that will allow me to merge as soon as I have the tests passing.