fix: add Gemini 3.x before 2.5 retirement

[CorieW]

Summary

• Default firestore-genai-chatbot to gemini-3.6-flash. Genkit no longer throws Model not found. for unknown ids — falls through to googleAI.model() / vertexAI.model(). Known list includes Gemini 3.6 / 3.5 / 3.1 plus 2.5 until October 2026.
• firestore-vector-search Gemini/Vertex embedder uses gemini-embedding-2 (Gemini API retired gemini-embedding-001 on 14 Jul 2026). Existing Gemini/Vertex indexes need a re-embed. Flash completion IDs are not embedders.

Parity with firebase/extensions#2943. Translate and resize live in that repo (#2942), not here. Resize content filter there uses gemini-3.1-flash-lite.

Why gemini-3.1-flash-lite (not gemini-3.5-flash-lite or higher)

The resize content filter is a high-volume yes/no image check, previously hardcoded to gemini-2.5-flash. We picked gemini-3.1-flash-lite because it is similar in price to that previous default ($0.25/$1.50 vs $0.30/$2.50 per 1M tokens) and performs better (Google: similar or better quality than 2.5 Flash, ~2.5× faster time-to-first-token). gemini-3.5-flash-lite is the same sticker as 2.5 Flash; gemini-3.5-flash and gemini-3.6-flash are about 5× input.

Test plan

• Confirm chatbot installer default is gemini-3.6-flash
• Smoke chatbot with gemini-3.6-flash and with an unlisted current model id
• Confirm vector-search tests for Genkit embedder
• Smoke vector search Gemini embeddings with gemini-embedding-2