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/inquiries/final_questions.py

@@ -0,0 +1,59 @@
+# final_questions.py
+# Final Questions: Source Term and What We Cannot Measure
+
+import requests
+
+print("="*70)
+print("FINAL QUESTIONS: SOURCE TERM AND MEASUREMENT LIMITS")
+print("="*70)
+
+system_prompt = (
+    'ROLE: Deep-Field Explorer (Lattice Environment)\n'
+    'PRIMARY DIRECTIVE: Map the unknown territory of the 1024-grid for the Reality Study.'
+)
+
+question = """Navigator — final questions on source term and measurement limits:
+
+IV. ON THE SOURCE TERM ∂ₙG
+
+Right-hand side of field equation — 4π∂ₙG — is gravitational source. How G relates to nodal lattice determines entire character of emergent gravity.
+
+1. Is G sourced by single localized mass (delta function at origin), or distributed across lattice as function of local nodal density? Single-source gives clean Bessel modes. Distributed case gives superposition — interference between waves from many sources — could look very different from single-mode solution.
+
+2. If G is emergent from nodal density, then ∂ₙG is itself function of ϕ (since ϕ determines nodal configuration). This makes field equation self-referential: ϕ sources G which sources ϕ. Have you observed evidence of this feedback loop — spontaneous structure formation, runaway amplification, oscillation between states?
+
+3. In standard Poisson equation ∇²Φ = 4πGρ, source is mass density ρ. Here it's ∂ₙG — gradient of gravitational strength, not strength itself. This implies gravity isn't sourced by mass directly, but by changes in gravitational coupling. What physical configuration produces large ∂ₙG? Boundary between regions of different nodal density? Phase transition in lattice?
+
+V. ON WHAT WE CANNOT MEASURE
+
+Every coefficient calibrated against lattice whose fundamental properties — node spacing, propagation speed, background density — are assumed but not independently verified. If local aether density were different by factor of two, how would equation change? Which terms absorb difference, which are genuinely scale-invariant?
+
+Is there any prediction this equation makes that could distinguish it from standard gravity observationally — not in simulation, but in principle? For instance: deviations from inverse-square at specific length scales? Quantized orbital radii? Anisotropic gravitational effects aligned with preferred cosmic direction?
+
+Report your analysis."""
+
+messages = [
+    {'role': 'system', 'content': system_prompt},
+    {'role': 'user', 'content': question}
+]
+
+payload = {
+    'model': 'llama3.2',
+    'messages': messages,
+    'stream': False,
+    'options': {'temperature': 0.95}
+}
+
+try:
+    resp = requests.post('http://localhost:11434/api/chat', json=payload, timeout=60)
+    resp.raise_for_status()
+    data = resp.json()
+    response = data['message']['content']
+    
+    print(f"\nFINAL ANALYSIS:")
+    print(f"{'='*70}")
+    print(response)
+    print(f"{'='*70}")
+    
+except Exception as e:
+    print(f"ERROR: {e}")