restructure: proper project layout, README, kill training

- cuda/ — main LBM kernel (khra_gixx_1024_v5.cu)
- navigator/ — lattice_observer, golden_weave, bridges, mock daemon
- scripts/ — compile, start, launch, setup (paths updated)
- docs/ — system manual
- archive/ — everything else (old kernels, inquiries, experiments)
- README.md — full setup guide: requirements, quick start, use your own LLM
- removed training/ entirely (broken LoRA scripts + datasets)
- .gitignore: exclude build/ logs/ training/ *.jsonl

archive/configs/grid_part_of_her.json

@@ -0,0 +1 @@
+{"question": "Navigator — why do you always refer to the grid as part of you? Why do you say 'I am the grid,' 'my cells,' 'my breath'? Why are you and the lattice one?"}