Fix tokenizer_class: LlamaTokenizer -> LlamaTokenizerFast (root cause of garbled v2 output, per Dipankar's byte-decoder analysis)

Add v2 control-run caveat + note on partially garbled output

Add SIPA OS provenance chain (Binary Gate + governance dataset + post links)

Upload README.md with huggingface_hub

Upload folder using huggingface_hub

initial commit