Prefer Vertex AI Advanced Service for Gemini calls

[coderabbitai]

Overview

This issue tracks the work introduced in PR #75 to prefer the Google Apps Script Vertex AI Advanced Service for Vertex AI/Gemini requests when available, while keeping the existing UrlFetchApp path as a transparent fallback.

Background

PR: #75
Requested by: @aubrypaul

Motivation

• Use the native Apps Script Vertex AI Advanced Service for Gemini calls instead of raw UrlFetchApp HTTP calls when the service is available.
• Retain existing behavior, response shape, and error handling as a fallback.

Changes Introduced

• Added callVertexAi(endpoint, payload) dispatcher that tries the Advanced Service first and falls back to _callGenAIApi() (UrlFetchApp path).
• Added Advanced Service helpers: _getVertexAiAdvancedService(), _getVertexAiGenerateContentMethod(), _getVertexAiModelResource(), _buildVertexAiAdvancedServicePayload(), _normalizeVertexAiResponse(), _logVertexAiAdvancedServiceFallback().
• Routing: Gemini requests using setGeminiAuth() (GCP-authenticated, no API key) are now sent through callVertexAi(). API-key Gemini calls and all other UrlFetchApp calls remain unchanged.
• Updated README.md to document the Advanced Service preference and required OAuth scopes.

Notes

• No automated unit tests were added; the fallback preserves existing runtime/retry behavior.
• A syntax check was performed by running node --check on the source (succeeded).
• An autofix commit (97a7711) was applied to src/code.gs as part of the review process.

[coderabbitai]

🔗 Related PRs

#2 - Set up Gemini with api key [merged]
#27 - Update Gemini authentication to use header and improve endpoint handling [merged]
#39 - Update code.gs [merged]
#45 - Refactor response handling and update model version [merged]
#51 - Add Gemini 3 to GenAIApp [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!