fix: correct COI formulation to measure price erosion over time

The fundamental error was treating COI as instantaneous margin × alpha.
The corrected formulation is:

    COI = E[p_start] - p_transaction

This measures price erosion over time, capturing how agents using
multiple sessions gather information and drive prices down.

Key changes:
- Add coi.py with COIWindow, COITracker, and compute_multi_session_coi
- Add separability.py with KL-divergence behavioral classification
- Update simplified_env.py to track initial prices and compute windowed COI
- Add corrected COI metrics (coi_*_corrected) alongside legacy metrics

The new approach:
1. Tracks prices at episode start as E[p] (expected price)
2. Computes transaction prices as p (actual sale price)
3. Measures leak as the difference (price erosion)
4. Includes order statistic erosion (Theorem 1: N agents -> min price)

lab/case/thesis/simplified.py

@@ -8,6 +8,14 @@ Objects:
 - Demand proxy q_hat via weighted action aggregation
 - COI leakage penalty for agent reconnaissance
 - Limbo: alternating price/demand history for trajectory analysis
+
+COI Correction (Jan 2026):
+The fundamental COI formulation is:
+    COI = E[p_start] - p_transaction
+
+This measures price erosion over time, not instantaneous margin × alpha.
+Agents use multiple sessions to gather information and find minimum prices.
+The price path from episode start to transaction captures information leakage.
 """
 from __future__ import annotations
 from dataclasses import dataclass, field