Keep summed chains in one row with their feeders tucked under

[BradyAJohnston]

Chains summed through a node's first input (e.g. the Add chains in MolecularNodes' Composite ID Outline / Cone Shadow, BradyAJohnston/MolecularNodes#1240) were laid out as a staircase, with all the feeders piled into the first columns.

• placement.py: when a node has two median predecessors, align with the one across the heavier link first, so a Float-only chain stays one row in all four BK runs.
• feeders.py (new): a node whose links out all go to one node of its frame is moved to the column just before that node (move_feeders_right), placed next to the other node of that column feeding the same node (tuck_feeders), and pulled against it after placement (pull_up_feeders).

Corpus snapshots updated; new summed_samples case and tests for it and value_chain. Net corpus cost goes up (more long links, e.g. Scatter on Surface crossings 241 -> 300), and the steps also run under "simple". To review more broadly before merging.

Before (main) above, after below:

Essentials

Images live on the pr-images/layout-feeder-rows branch; delete it after merging.

[codecov]

Codecov Report

❌ Patch coverage is 98.79518% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 98.31%. Comparing base (3f0481c) to head (fd378ef).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/nodebpy/layout/feeders.py	98.18%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #254      +/-   ##
==========================================
+ Coverage   98.29%   98.31%   +0.01%     
==========================================
  Files         105      106       +1     
  Lines       31104    31168      +64     
==========================================
+ Hits        30575    30642      +67     
+ Misses        529      526       -3     

☔ View full report in Codecov by Harness.
📢 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.

[codspeed]

Merging this PR will improve performance by 38.14%

⚠️ 1 benchmark spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• test_build_decoder_4bit

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 3 improved benchmarks
✅ 12 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	test_arrange_decoder_4bit[sugiyama]	5.3 s	3.5 s	+51.16%
⚡	test_arrange_math_expression	2.4 s	1.8 s	+35.41%
⚡	test_arrange_decoder_4bit[simple]	1.7 s	1.3 s	+28.78%

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing layout-feeder-rows (fd378ef) with main (3f0481c)¹}

Footnotes

1. No successful run was found on balance-socket-ids (6e78b4d) during the generation of this report, so main (3f0481c) was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩