feat(ml): replace logistic regression with MLP fusion and KS drift with ADWIN online learning

Replace the LogisticRegression meta-learner with a PyTorch MetaFusionMLP
(Linear(3,16)->BN->ReLU->Dropout->Linear(16,1)->Sigmoid) for non-linear
fusion of EIF, NF, and XGBoost scores. Replace KS-test + quantile digest
drift detection with ADWIN (adaptive sliding window, Hoeffding bound).
Replace weekly XGBoost batch retraining with River HoeffdingAdaptiveTree
for incremental online learning (learn_one per cycle). Update all thesis
documentation sections (2.4.2c, 2.4.3, 3.8, discussion, conclusion).

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

services/bot-detector/bot_detector/requirements.txt

@@ -10,5 +10,6 @@ torch_geometric>=2.4
 FrEIA>=0.2
 xgboost>=2.0
 cleanlab>=2.6
+river>=0.19
 pyyaml>=6.0
 ja4-common @ file:///app/shared/ja4_common