fix: add Gemini 3.x before 2.5 retirement

[CorieW]

Summary

• Add Gemini 3.6 Flash, 3.5 Flash, 3.5 Flash Lite, 3.1 Flash Lite, and 3.1 Pro Preview to firestore-translate-text and default to 3.6 Flash. Gemini 2.5 options remain until they retire in October 2026.
• Switch storage-resize-images content filter from gemini-2.5-flash to gemini-3.1-flash-lite.
• Both Vertex AI call sites now target the global endpoint instead of the Cloud Functions location — the translate Genkit client and the resize content filter.
• Bump firestore-translate-text to 0.1.31 and storage-resize-images to 0.3.7, with CHANGELOG and generated README updates.

Vertex AI location

gemini-3.6-flash and the other Gemini 3.x models are only served from the Vertex AI global, us and eu endpoints (locations); a single region such as us-central1 returns a 404. Translate passed process.env.LOCATION into the Vertex plugin, so a Vertex install on the new default model would have failed. It now uses global, matching the content filter. Neither extension exposes a Vertex location param, so this is hardcoded with a comment rather than added to the installer.

The old @genkit-ai/vertexai plugin rejects global, which is why the content filter moves to @genkit-ai/google-genai.

Sampling controls

Not applicable here — neither firestore-translate-text nor storage-resize-images exposes TEMPERATURE, TOP_P or TOP_K. (Gemini 3.x deprecates those controls; that is handled in the chatbot extensions.)

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 translate installer lists 3.x models and defaults to gemini-3.6-flash
• Confirm 2.5 models still selectable for existing installs
• Run firestore-translate-text unit tests and tsc --noEmit
• Run storage-resize-images content-filter unit tests
• Smoke a Gemini translation with the new default model on Vertex at global
• Smoke resize with content filter enabled (Vertex global; regional us-central1 404s this model)
• Regenerate firestore-translate-text/README.md via npm run generate-readme

[gemini-code-assist]

Code Review

This pull request updates the Gemini models used in the firestore-translate-text and storage-resize-images extensions. Specifically, it introduces Gemini 3.6 Flash, 3.1 Flash Lite, and 3.1 Pro Preview model options, setting Gemini 3.6 Flash as the new default model. It also updates the content filter in the image resizing extension to use Gemini 3.6 Flash, and updates the corresponding documentation, configuration files, and tests. There are no review comments, so I have no feedback to provide.