update cloudflare-workers-ai: add gemma-4, remove non-LLMs, fix metadata

- Add gemma-4-27b-a4b-it (multimodal, reasoning, tool calling)

- Remove non-LLM models: embeddings, TTS, translation, sentiment analysis

- Remove redundant models: Llama 2/3.x variants, Qwen, Mistral, DeepSeek, Gemma 3

- Update metadata: tool_call, reasoning, open_weights, attachment for remaining models

- Final models: gemma-4, llama-4-scout, kimi-k2.5, nemotron-3, gpt-oss-20b/120b, glm-4.7-flash

Author: mchenco