[TASK] dotAI: LangChain4J - Google Vertex AI provider

[fmontes]

Description

Add Google Vertex AI (Gemini) as a supported chat provider in the dotAI LangChain4J integration layer. Authentication is handled via Application Default Credentials — no API key required.

This builds on the LangChain4J abstraction established in #35183 (Phase 1 — OpenAI).

Implementation outline:

• Add langchain4j-vertex-ai-gemini dependency (with enforcer exclusions for org.checkerframework:checker-qual and com.google.android:annotations)
• Add vertex_ai case to LangChain4jModelFactory switch
• Implement buildVertexAiChatModel using VertexAiGeminiChatModel
• buildVertexAiEmbeddingModel and buildVertexAiImageModel throw UnsupportedOperationException

Acceptance Criteria

• vertex_ai is selectable as a chat provider in providerConfig
• Authentication uses ADC automatically; no API key is accepted
• projectId and location are required configuration fields
• Embeddings and image generation throw UnsupportedOperationException with a clear message
• Unit tests cover Vertex chat builder and unsupported-operation paths in LangChain4jModelFactoryTest
• Maven enforcer exclusions are scoped only to the Vertex dependency

Additional Context

• Parent: [FEATURE] dotAI: LangChain4J integration — Phase 1 (OpenAI) #35183 (Phase 1 - OpenAI)
• Epic: EPIC: dotAI Multi-Provider Support #33970 (dotAI Multi-Provider Support)
• Implementation PR: feat(dotAI): Dot AI LangChain4J - Google Vertex #35241

Sample configuration:

{
  "chat": {
    "provider": "vertex_ai",
    "projectId": "my-gcp-project",
    "location": "us-central1",
    "model": "gemini-1.5-pro",
    "maxTokens": 8192,
    "temperature": 1.0
  }
}

The host running dotCMS must have a valid GCP identity (gcloud auth application-default login or an attached service account).