fix: restore Qwen/Qwen3.5-2B tokenizer_config (transformers 4.x compatibility; v5 save artifacts removed)

fix: tokenizer_class TokenizersBackend -> Qwen2Tokenizer (transformers 4.x compatibility)

Update README.md

Update README.md

Update README.md

docs: add explanation-variant prompt and structured-output tip for pure-prompt integrations

Add LettuceCode mascot + tagline

Card: tested transformers example + comparison tables + TL;DR

Card: TL;DR results, RAGTruth/PsiloQA comparison tables, LFM/gpt-oss rows

lettucedect-v2-qwen-2b: generative code+prose hallucination span detector

initial commit