Layer 1.5: architecture-general neural existence proof

Re-scopes Layer 2 into a cheaper, architecture-general neural collapse proof
before the LLM rung. Realises the same Wright–Fisher abstractions in real trained
generative models on a fully-synthetic sandbox with an exact oracle, reusing
knowledge.metrics/truth/seeding and the output contract so neural curves overlay
the Layer-1 analytic curves.

  - src/neural/: synthetic token-grammar sandbox (lossless identity + stochastic
    style), ExactOracle, HistogramModel bridge, generation loop, experiment runner
  - HARD GATE passed: histogram lineage reproduces Layer 1 exactly (neutral decay,
    exact H_eq, tracks run_lineage) — tests/test_neural_validation.py
  - torch models: autoregressive RNN + MLP (VAE implemented, not yet fidelity-
    passing); determinism seeding derived from the SeedSequence stream
  - N0 bridge (neural g*=0.047 ≈ Layer-1 0.048), N1 collapse-in-weights, N2 phase
    boundary, N5 architecture-generality (collapse + grounding-rescue in histogram
    + RNN + MLP). Manifests/configs committed; parquet gitignored, hashes tracked
  - additive backward-compatible save_artifacts extension; Makefile neural targets

Finding: neural smoothing partially resists H-collapse, so forward-KL and tail
survival are the sharp neural collapse metrics (H is smooth, per Layer 1).

92 tests green. Remaining (tasks/todo.md): N4 merge, N2 refine, N3/N6, VAE
fidelity, MNIST tier, figures. LLM/LoRA rung and C3 deferred.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.gitignore

@@ -1,13 +1,8 @@
-# Environments
+# Reproducibility: results artifacts are regenerable from committed configs + seeds; only
+# their hashes (in each run's manifest.json) are tracked, per the open-science contract.
 .venv/
 __pycache__/
 *.pyc
 .pytest_cache/
-
-# Results: the large data artifact (results.parquet) is regenerable and gitignored;
-# the reproducibility metadata (manifest.json with content hashes, resolved_config.yaml)
-# and the figures (PNG/PDF) are tracked so the paper's figures live in the repo.
-results/**/*.parquet
-
-# OS / editor
-.DS_Store
+results/**/results.parquet
+models/