ENH: tpx 138 (GROMACS 2026.0) support

[tylerjereddy]

• Add .tpr parsing support for GROMACS 2026.0, which was recently released:
  https://manual.gromacs.org/2026.0/release-notes/2026/major/highlights.html

• I believe this is the first time we bump the tpx version support since we added support for parsing positions and velocities from .tpr files, so we may want to be particularly careful here.

@al42and I just used gmx convert-tpr with GROMACS 2026.0 on your alanine dipeptide NN potential .tpr provided back in gh-4919. Will that suffice to test out the new tpr "bells and whistles" from GROMACS 2026? The release highlights mention "Expanded support for running simulations with Neural Network Potential models, now including link atom treatment for NNP/MM, pairlist input, and electrostatic embedding models." But for the most part the .tpr parsing doesn't seem to need any new shims with the .tpr files we're currently using.

LLM / AI generated code disclosure

LLMs or other AI-powered tools (beyond simple IDE use cases) were used in this contribution: No

PR Checklist

• Issue raised/referenced?
• Tests updated/added?
• Documentation updated/added?
• package/CHANGELOG file updated?
• Is your name in package/AUTHORS? (If it is not, add it!)
• LLM/AI disclosure was updated.

Developers Certificate of Origin

I certify that I can submit this code contribution as described in the Developer Certificate of Origin, under the MDAnalysis LICENSE.

---

📚 Documentation preview 📚: https://mdanalysis--5219.org.readthedocs.build/en/5219/

[tylerjereddy]

note: I didn't actually check the tpx generation (number 28 here), so we may want to double check that

[IAlibay]

If I'm following this enum properly, we're on generation 29? https://github.com/gromacs/gromacs/blob/main/src%2Fgromacs%2Ffileio%2Ftpxio.cpp#L244

(generation is Count-1)

[tylerjereddy]

Note: these new .tprs were all produced using gmx convert-tpr with GROMACS built from source at the release tag for version 2026.0.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.72%. Comparing base (24548e6) to head (2270ac5).

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #5219      +/-   ##
===========================================
- Coverage    92.73%   92.72%   -0.01%     
===========================================
  Files          180      180              
  Lines        22475    22475              
  Branches      3190     3190              
===========================================
- Hits         20842    20841       -1     
- Misses        1176     1177       +1     
  Partials       457      457              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

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

[IAlibay]

As far as I understand, this looks good to me, just the one tiny thing.

[IAlibay]

Suggested change

	138 28 2026.0 yes
	138 29 2026.0 yes

[tylerjereddy]

I used the suggested changes GUI interface to apply the one suggested documentation change based on reviewer feedback (thanks for checking that.. I'm swamped).

[al42and]

I just used gmx convert-tpr with GROMACS 2026.0 on your alanine dipeptide NN potential .tpr provided back in #4919. Will that suffice to test out the new tpr "bells and whistles" from GROMACS 2026?

In 2027, we consider embedding NN weights into TPR, which can get funky, but for now it's just a couple new fields.
Your approach should be fine unless you aim for 100% coverage :)