fix: _as_index() calls to_index() for DataArrays

[peterhron]

Closes #898 (if applicable).

Changes proposed in this Pull Request

helper in alignment.py uses DataArray.to_index() to retrieve the actual pandas index object instead of re-building it via pd.Index. This fixes regression where timezone-aware indexes cannot be used to build a model.

Unit tests are added to validate that models can be built with timezone-aware coords.

Checklist

• AI-generated content is marked (see AGENTS.md).
• Code changes are sufficiently documented; i.e. new functions contain docstrings and further explanations may be given in doc.
• Unit tests for new features were added (if applicable).
• A note for the release notes doc/release_notes.rst of the upcoming release is included.
• I consent to the release of this PR's code under the MIT license.

[codspeed-hq]

Merging this PR will regress 2 benchmarks

⚡ 6 improved benchmarks
❌ 2 regressed benchmarks
✅ 167 untouched benchmarks
⏩ 175 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Memory	test_to_lp[storage-n=250]	30.3 MB	35.8 MB	-15.43%
❌	Memory	test_to_lp[nodal_balance-severity=50]	2.9 MB	3.3 MB	-12.05%
⚡	Memory	test_to_lp[nodal_balance-severity=0]	3.4 MB	2.4 MB	+41.85%
⚡	Memory	test_to_lp[milp-n=50]	2.6 MB	2 MB	+29.32%
⚡	Memory	test_build[knapsack-n=100]	5.5 KB	4.7 KB	+17.07%
⚡	Memory	test_to_lp[merge_balance-severity=0]	2.6 MB	2.3 MB	+15.59%
⚡	Memory	test_build[knapsack-n=10000]	830.3 KB	752.2 KB	+10.39%
⚡	Memory	test_to_lp[storage-n=10]	2.9 MB	2.6 MB	+10.17%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing peterhron:fix/timezone-aware-coords (33a2f82) with master (f06738e)}

Footnotes

1. 175 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[peterhron]

@codspeedbot fix this regression

[peterhron]

hey, @FBumann @FabianHofmann , tagging you guys since you were the last to touch the related code; could you please briefly check this? Details are in the linked issue, pls let me know whether you find this resolution appropriate.

If so, i'd like to ask for brief assistance with fixing the checks, as this is my first contribution here and i'm not familiar with it. Both the failures are weird to me, and likely transient (but i can't re-run the pipelines to confirm).

I'd think the memory regression is just noise (codspeed also reports 'Different runtime environments detected', which might be the reason). The test failure on 3.11. is unrelated to my changes - Copilot suggested that it's a problem with a flaky version of coptpy. Similar error happened on master today.

Pleaes let me know what you think. Thanks a lot!

[FBumann]

Hi @peterhron,
First of all thanks for your Pr. It looks very reviewable and focused to me. We'll get it reviewed soon.

About Codspeed: Yes, Codspeed is quite flaky atm. This won't be a regression.

About the tests: Yes, this isn't from your change. We are investigating an issue with COPT at the moment #899

[FBumann]

Great work! Please add a release note entry, then we are good to go @peterhron.

[FBumann]

Thanks for your contribution @peterhron

[peterhron]

Thanks for your contribution @peterhron

thanks for the quick response. I just had to rebase, otherwise the PR should be good to go!