fix: pre-translate non-English chunks to English before LLM inference

fix: enable sampling — temperature=0.8 for MCQ, 0.4 for evaluator

fix: stronger MCQ/evaluator prompts — no duplicates, no self-reference, English-only, section 2 always present

fix: make the llama.cpp CPU runtime actually usable on the Space

fix: truncate runaway question generation to the first question

feat: amphitheater hero with mouse-tracking robot professor, fixed particles

fix: force English output in all LLM prompts regardless of PDF language

fix: server error

fix: increase GPU duration to 120s, prefetch FLUX weights on startup

fix: increase GPU duration to 120s, prefetch FLUX weights on startup

fix: FLUX.2-klein reajust

fix: FLUX.2-klein prompt keyword arg, image for all verdicts

feat: switch to FLUX.2-klein-4B for image generation

feat: FLUX.1-schnell image generation after correct answers

feat: add MCQ mode with 4 options, client-side evaluation and per-choice explanations

feat: Off-Brand badge - custom FastAPI+HTML frontend, zero Gradio chrome

fix: prompts évaluateur séparés EN/FR pour forcer la langue du feedback

fix: texte question/feedback invisible + prompt langue renforcé

fix: tighten question prompt to avoid compound multi-part questions

feat: switch to MiniCPM4-8B, add robustness, score tracker, language selector

feat: initial project structure with all core modules