feat: Agentic Positive Prompting & Scenario Tests

Refactor: Modularized codebase and consolidated Oracle bridges

Fix Role Alternation crash for assistant-led conversations and extend simulation tests

Sage 6.5: Transform into lean text-only agent, remove RAG/Voice/Mongo, optimize model loading, and extend UI/Logic verification tests

fix: agent history integration, OOM protection, and UI cleanup

feat: remove MongoDB and Sacred Knowledge tab, add agent welcome message, fix typos

fix: Use AutoModelForCausalLM for Gemma 3 1B (text-only) to fix loading issues

feat: Explicit AutoProcessor/Tokenizer logic and verified 1B/4B/4B-Quant tool calling

refactor: Modularize app structure (app_module.py + thin entry points)

feat: Add support for 'name' argument in Oracle tool calls

fix: Improve conversation fluidity with prompt injection and verified filtering

feat: Agentic Oracle, Multi-Message Bubbles, Filtering, and UI Fixes