KCL: Fix do_post_extrude querying wrong body for clones/mirrors

[adamchalmers]

Terminology note: old body = before clone, new body = after clone

4 problems here:

1. The logic for choosing the right edge to use in the API queries (i.e. choosing which edge ID to assign to extrusion_info_edge_id) was very confusing to follow, due to mutation. I refactored it to be immutable, and added comments explaining its logic.
2. The logic was wrong: it was sometimes using an ID from the new body, when it's supposed to query the old body.
3. The get_adjacency_info call was not using the extrusion_info_edge_id, it was using any_edge_id, which is one of the inputs to the complex logic above, not its final output after carefully analyzing which edge ID to use.
4. the get_adjacency_info call was using the wrong sketch: sketch_id not sketch.id. I hate that these are two different things.

The broader takeaway is that do_post_extrude sucks, I hate this function and Serena and I have a plan to kill it (see KittyCAD/modeling-api#1130)

Testing

Currently, I've tested it by running a local engine and looking at the logs. Before this, the engine logs Unexpected missing opposite edge encountered during Solid3dAdjacencyInfo::get() from Mike's recent PR. As of this PR, the logs no longer show up.

You can also see the sim test for "christmas_tree_mirror3d_union" has been updated and adds 3 "SweepEdge Adjacent" entries to the artifact graph.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
modeling-app	Ready	Preview, Comment	Jun 12, 2026 5:07am

[codspeed-hq]

Merging this PR will degrade performance by 26.59%

❌ 1 regressed benchmark
✅ 124 untouched benchmarks
🆕 1 new benchmark
⏩ 151 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	parse_i-beam	16 ms	21.8 ms	-26.59%
🆕	Simulation	parse_artificial-heart	N/A	185.1 µs	N/A

Tip

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

---

_{Comparing achalmers/try-the-fn-test-again (27e0177) with main (36d2e6a)²}

Footnotes

1. 151 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 (b4c3fbf) during the generation of this report, so 36d2e6a was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[jtran]

Nice!

[jtran]

Fixes #11994.