Triage: merge or close 9-month-old PR #4 and prune stale branches

[Zychs]

Context

PR #4 ("Hook up ternary snake prototype") has been open since 2025-11-16 — over 9 months. It targets godot (not main), is mergeable (clean state), has no CI checks, and was never tested in-engine.

Stale branches:

• codex/implement-tertiary-movement-vectoring — already merged via PR Implement ternary snake movement prototype #2
• codex/implement-tertiary-movement-vectoring-8rwxuw — PR Hook up ternary snake prototype #4 head, untested
• GMS — purpose unknown
• non-euclidean-prototype — purpose unknown
• godot — intermediate branch, already merged to main via PR Godot #3

Decision needed

1. Merge PR Hook up ternary snake prototype #4 into godot → then merge godot into main (or close if the approach is superseded)
2. Delete already-merged branches (codex/implement-tertiary-movement-vectoring, godot post-merge)
3. Decide on GMS and non-euclidean-prototype: keep or archive

Why now

Zombie PRs and orphan branches are cognitive load. Either the ternary prototype is worth keeping or it isn't — 9 months of silence is the answer if you let it be.