updated readme, ruff formatted all files

Cleaned up code, added multiseed training wrapper, PyTorch profiler training option, updated gradio demo, made changes to research paper to match new changes and new training results from adding new training techniques, architecture.md now explains all designs and decisions

Added many new improvements based on feedback from others

Full training results & evaluation with BERTScore

Clean up codebase and fix training bugs

Full training run, code cleanup, mypy/ruff fixes

feat: Add FLAN-T5 compatibility with relative position bias

Refactor: Fix B905 zip() strictness and resolve mypy error in evaluate.py

Style: Apply ruff formatting

Refactor: Consolidate dependencies, improve testing, and add CI/CD

chore: snapshot current refinements