ci(benchmarks): a published result records specsolve's release version rather than 0.0.1

[FBumann]

Prompt: Yes, close it and fix the version

Note

The following content was generated by AI.

The measuring job checked out one commit with no tags, so every published result records specsolve as 0.0.1.dev1+g<sha>. fetch-depth: 0 on that checkout gives the version counted from the nearest tag.

Cause, verification, and what is not done

The eight files on main from #2039 all say 0.0.1.dev1+gfbc735a5b; the lock installs 0.9.1.dev31+gfbc735a5b. The version comes from hatch-vcs ([tool.hatch.version] source = "vcs"), which needs the tags and the history back to the nearest one.

Reproduced on main at 61172a8, with setuptools-scm:

checkout	version
depth 1, no tags (what actions/checkout does by default)	0.1.dev1+g61172a8af
full history and tags	0.9.1.dev34+g61172a8af

• The other checkouts in this workflow and ci.yml already use fetch-depth: 0.
• Ran: tests/test_architecture.py -k workflow, 13 passed. Not run: pixi run check, and a real run on the box.
• The lock test was not affected, because _same_install matches on the commit and ignores the release number. So nothing failed and nothing here changes it.
• Not done: the eight result files on main keep 0.0.1.dev1 until the next published run. Measured files are not edited by hand.
• The branch carries a merge of its earlier head fbc735a (closed test(ci): the published-benchmark workflow parses again, and the suite refuses a workflow that does not #2038, already in main through docs(benchmarks): the published benchmarks are re-measured at the new model sizes, five times each tenfold step #2039) so the push was a fast-forward; the diff against main is the one line.

Type ci: the diff touches only .github/, so no changelog line.

🤖 Generated with Claude Code

https://claude.ai/code/session_01NTByC5UJJWvcFj3mvmv83v

---

Generated by Claude Code

[codspeed]

Merging this PR will not alter performance

✅ 104 untouched benchmarks
⏩ 307 skipped benchmarks¹

---

_{Comparing claude/specsolve-2032-review-7hrcv0 (755fc7c) with main (2588883)²}

Footnotes

1. 307 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. ↩

2. No successful run was found on main (61172a8) during the generation of this report, so 2588883 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩