FCE-2750 Add support for agent image capture

[Qizot]

Description

Describe your changes.

Motivation and Context

Why is this change required? What problem does it solve? If it fixes an open
issue, please link to the issue here.

Documentation impact

• Documentation update required
• Documentation updated in another PR
• No documentation update required

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to
  not work as expected)

[linear]

FCE-2750 Add support for image capture in python sdk

[Copilot]

Pull request overview

This PR adds support for agent image capture functionality, enabling agents to request and receive images from video tracks in real-time. The changes include new protobuf message types for image capture requests/responses and a multimodal example demonstrating integration with Google's Gemini Live API.

Changes:

• Added capture_image method to AgentSession for requesting images from video tracks
• Introduced IncomingTrackImage message type for receiving captured images
• Created a comprehensive multimodal example showing audio+video interaction with Gemini Live API

Reviewed changes

Copilot reviewed 17 out of 19 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
pyproject.toml	Added multimodal example to workspace members
protos	Updated submodule to include agent image capture protocol definitions
fishjam/integrations/gemini.py	Added metadata to output audio track settings
fishjam/events/_protos/fishjam/init.py	Added protobuf message types for image capture and channel management
fishjam/agent/agent.py	Implemented capture_image method and IncomingTrackImage message handling
fishjam/agent/init.py	Exported IncomingTrackImage type
examples/multimodal/pyproject.toml	Configuration for new multimodal example project
examples/multimodal/multimodal/worker.py	Background task worker for managing async operations
examples/multimodal/multimodal/session.py	Gemini Live API session management for audio and image streaming
examples/multimodal/multimodal/room.py	Fishjam room service managing room and agent lifecycle
examples/multimodal/multimodal/notifier.py	Event handlers for peer and track lifecycle notifications
examples/multimodal/multimodal/config.py	Configuration for Gemini model and capture settings
examples/multimodal/multimodal/agent.py	Multimodal agent coordinating video capture and Gemini interaction
examples/multimodal/main.py	FastAPI application entry point
examples/multimodal/README.md	Documentation for the multimodal example
compile_proto.sh	Updated protobuf compilation to use uv run
.gitmodules	Updated protos submodule to use agent-capture-image branch

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.