Tighten paper, clean README, add .gitignore

Cut ~40% paper text — tables speak for themselves.
Remove duplicate episodic memory framing from README.
Add serve.py to quick start.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.gitignore

@@ -1,2 +1,5 @@
 models/*.onnx
 models/*.onnx.data
+__pycache__/
+*.pyc
+paper.out