Add GCP Vertex AI model provider support

[ystarikovich]

Summary

langmem currently ships with langchain-openai and langchain-anthropic as dependencies, but has no built-in support or documentation for using GCP Vertex AI models via langchain-google-vertexai.

Many production workloads run on GCP and use Vertex AI as their LLM gateway. Since langmem's internal functions use init_chat_model() from langchain, Vertex AI models should work when passing a string like "google_vertexai:gemini-2.0-flash" — but this is neither documented nor tested.

Request

• Add langchain-google-vertexai as an optional dependency:

  [project.optional-dependencies]
  gcp = ["langchain-google-vertexai>=2.0.0"]

• Add documentation / examples showing how to use Vertex AI models with langmem
• Validate that core features (memory extraction, prompt optimization, etc.) work with ChatVertexAI

Example usage

from langchain_google_vertexai import ChatVertexAI
from langmem import create_memory_store_manager

model = ChatVertexAI(model="gemini-2.0-flash")
manager = create_memory_store_manager(model)

or via string identifier:

manager = create_memory_store_manager("google_vertexai:gemini-2.0-flash")