Added langgraph example

[cnemri]

Title: Add "Agentic Accessibility Checker" Demo

Description:
This pull request introduces a new demo to the AAIE samples repo: the Agentic Accessibility Checker. The demo provides an end-to-end example of building an autonomous agent for real estate accessibility analysis using Google’s cutting-edge AI tools, including Gemini and the Vertex AI Reasoning Engine. It includes:

• Synthetic Data Generation: Utilizes generative models to create realistic multimodal real estate data (text and images), stored in GCS and Firestore.
• Cloud Run API: A FastAPI-based service acting as the agent’s environment for retrieving property details.
• LangGraph Agent with Gemini: A ReAct-style agent that can reason about and interact with data, tools, and APIs, leveraging both text and image understanding capabilities.
• Streamlit Frontend: A user-friendly interface that allows interactive exploration of properties and direct engagement with the agent.
• Comprehensive Setup Instructions: Detailed steps for installation, data generation, API deployment, agent deployment, and frontend setup, ensuring a smooth onboarding experience.
• Integration with Google Cloud Services: integrates with Firestore, GCS, and Vertex AI Reasoning Engine, includes environment-specific parameters.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[code-review-assist]

Hi team,

I'm currently reviewing this pull request and will post my full review in a few minutes. In the meantime, here's a quick summary to help you and other reviewers get up to speed:

This PR adds a new demo to the applied-ai-engineering-samples repository: the Agentic Accessibility Checker. This demo showcases a complete system for analyzing real estate accessibility using Gemini and the Vertex AI Reasoning Engine. Key features include synthetic data generation (using Gemini and Imagen3), a FastAPI-based Cloud Run API, a LangGraph agent (ReAct style), and a Streamlit frontend. The PR includes new Dockerfiles for the API and Streamlit app, updated requirements files, a comprehensive README, and deployment scripts. The core change is the addition of the src directory containing the agent, API, and Streamlit app code, along with supporting files for data generation and deployment.

If there's anything unclear or requires further clarification, please let me know.

---

And now, a little haiku to brighten your day:

Agent learns and grows,
Cloud's vast knowledge at its hand,
Access blooms anew.

[code-review-assist]

Gemini encountered an error creating the review. You can try again by commenting @code-review-assist review.

[RajeshThallam]

@code-review-assist review

[code-review-assist]

This pull request introduces a new LangGraph-based demo for analyzing real estate accessibility. The demo uses Gemini, Vertex AI Reasoning Engine, and several other Google Cloud services. The overall design and implementation are well-structured. The documentation is comprehensive and provides clear instructions. Here are some suggestions for improvement:

[RajeshThallam]

@code-review-assist review

[code-review-assist]

This pull request introduces a new LangGraph-based demo for analyzing real estate accessibility. The demo uses Gemini, Vertex AI Reasoning Engine, and a Streamlit frontend. The overall design and implementation are well-structured. The documentation is comprehensive and provides clear instructions for setup and deployment. However, there are a few minor improvements that can enhance the code's robustness and maintainability.