Improve aluminum model

[macflo8]

This PR adds the improvements of the aluminum sector representation in MESSAGEix-Materials developed in the course of the ScenarioMIP project. It cherry-picks some initial commits from the #235 branch.

How to review

• Read the diff and note that the CI checks all pass.
• Run MESSAGEix-Materials build and check if it runs successfully
• Look at the materials doc in the ReadTheDocs preview build of the documentation.
• Ensure that changes/additions are self-documenting, i.e. that another
  developer (someone like the reviewer) will be able to understand what the code
  does in the future.

PR checklist

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

[khaeru]

Thanks @macflo8!

The error message on CI is a little opaque, but I believe it occurs because one of the added/modified files wants to import .material.data_other_industry that is missing from not yet included in the PR.

[macflo8]

Yes, good catch. I had a conflict during rebasing and did not merge correctly it seems. I will try to repeat the step and clean that up.

[codecov]

Codecov Report

❌ Patch coverage is 89.89547% with 29 lines in your changes missing coverage. Please review.
✅ Project coverage is 73.4%. Comparing base (a7c26bd) to head (c4ce855).
⚠️ Report is 776 commits behind head on main.

Files with missing lines	Patch %	Lines
message_ix_models/model/material/data_aluminum.py	92.4%	17 Missing ⚠️
message_ix_models/model/material/build.py	0.0%	6 Missing ⚠️
message_ix_models/model/material/data_util.py	50.0%	6 Missing ⚠️

❗ There is a different number of reports uploaded between BASE (a7c26bd) and HEAD (c4ce855). Click for more details.

HEAD has 1 upload less than BASE

Flag	BASE (a7c26bd)	HEAD (c4ce855)
	16	15

Additional details and impacted files

@@           Coverage Diff           @@
##            main    #345     +/-   ##
=======================================
- Coverage   78.6%   73.4%   -5.3%     
=======================================
  Files        219     220      +1     
  Lines      17499   17746    +247     
=======================================
- Hits       13766   13034    -732     
- Misses      3733    4712    +979     

Files with missing lines	Coverage Δ
message_ix_models/model/material/data_generic.py	14.9% <ø> (-83.6%)	⬇️
message_ix_models/model/material/data_petro.py	11.1% <ø> (-67.8%)	⬇️
message_ix_models/model/material/util.py	58.5% <100.0%> (+14.6%)	⬆️
...ssage_ix_models/tests/model/material/test_build.py	92.3% <ø> (-3.9%)	⬇️
..._models/tests/model/material/test_data_aluminum.py	100.0% <100.0%> (ø)
message_ix_models/model/material/build.py	54.7% <0.0%> (-13.2%)	⬇️
message_ix_models/model/material/data_util.py	12.9% <50.0%> (-20.5%)	⬇️
message_ix_models/model/material/data_aluminum.py	93.4% <92.4%> (+1.8%)	⬆️

... and 13 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]

Thanks @macflo8 for the work and collaboration here!

All checks are passing (except one, see below), so approved.

Just to recap some points we discussed for follow-up:

• We accept the drop in coverage for the time being—to be addressed in a follow-up PR after other material pieces are on main.
• #348
• We can investigate how to replace .material.util.read_config() with a Config class, to avoid the Context.get_instance(-1) call that seems fragile on GitHub Actions/in the test suite. This should in turn allow to remove the one XFAIL test mark added in this PR.

I will now:

• Rebase the PR on latest main.
• Merge (without waiting for the checks to all run again).
• Rebase the ssp-dev branch of #235 on main.