feat(ml): replace Autoencoder with RealNVP Normalizing Flow and add SessionTransformer embeddings

Replace TrafficAutoEncoder (MSE reconstruction scoring) with TrafficNormalizingFlow
(RealNVP via FrEIA, 4 affine coupling blocks, anomaly score = -log p(x)) for
mathematically rigorous density estimation. Add SessionTransformer module producing
32-dimensional sequence embeddings from raw HTTP request sequences (path, method,
timing) via a lightweight TransformerEncoder, replacing path_transition_entropy and
cadence_cv features. Update thesis documentation sections 2.4.2b and 3.8 accordingly.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

services/bot-detector/bot_detector/config.py

@@ -79,6 +79,11 @@ AE_EPOCHS = int(os.getenv('AE_EPOCHS', '50'))
 AE_LATENT_DIM = int(os.getenv('AE_LATENT_DIM', '16'))
 AE_LEARNING_RATE = float(os.getenv('AE_LEARNING_RATE', '1e-3'))
 
+SESSION_TRANSFORMER_PATH = os.getenv(
+    'SESSION_TRANSFORMER_PATH',
+    os.path.join(MODEL_DIR, 'session_transformer.pt')
+)
+
 # ─── XGBoost — troisième voix supervisée ────────────────────────────────────
 
 XGB_WEIGHT = _require_float('XGB_WEIGHT', 0.20, 0, 1)