feat: calibrate account risk scores and threat tiers, resolve clumping in alert queue

fix: resolve investigation overlay stuck and add hybrid models to startup download

feat: add hybrid GNN/ML models and improved investigation UI

Data upload feature implemented

fix: serve landing page at root, add /app route for SPA

DevOps: dynamically download pagerank, betweenness, louvain and gnn metrics checkpoints from HF Hub on startup

DevOps: pass HF_TOKEN to hf_hub_download to enable authenticated downloads

DevOps: dynamically download graph_edges, features and alerts checkpoints from HF Hub on startup

DevOps: optimize repo by untracking models and dynamic data assets; implement blended alert risk calibration

fix: calibrate XGBoost risk scores & classify alert queue by tier

fix: restore Background.png to frontend/pages/ served at /pages/Background.png (fastest static path)

fix: serve Background.png via dedicated /bg.png route to bypass static mount issues

feat: merge Chauhan frontend enhancements and fix pinecone dependency

fix: remove simulated XGBoost metrics; harden SHAP against string dtypes

fix: detect and replace LFS pointer files at startup

feat: download models from HF Hub at startup; improve GNN trainer

fix: global NaN-safe JSON encoder on FastAPI app — covers all routes

fix: restore ML models, fix deployment paths, sync alert risk scores with XGBoost

feat(copilot): complete copilot integration and resolve CSS styling compliance

feat: integrate case management with Supabase authentication and storage

Prepare frontend for Cloudflare and Vercel deployment

feat: decouple frontend and fix investigation 500 error

deploy: add Dockerfile, vercel.json and update server.py for deployment

feat: Implement caching for API responses and SHAP explanations to improve performance and reduce redundant computations.

Fix investigation view rendering, fix STR PDF generation (unicode/limit), and add graph checkpoint for fast server restart

changed