docs: pipeline_tag image-text-to-text (was any-to-any)

feat: add BF16 + Q8_0 gguf to complete the quant set for paired evals

docs: library_name: gguf so Ollama HF pull integration picks up this repo

feat: add Q4_K_M gguf for Ollama / llama.cpp consumers

Update README.md

Add EUPL-1.2 license

Duplicate from google/gemma-4-E2B-it