[New Feature]: MMGIS Copilot agent using Azure AI service

[midatm1234]

Checked for duplicates

Yes - I've already checked

Alternatives considered

Yes - and alternatives don't suffice

Related problems

Summary

• Add “MMGIS Copilot” floating panel that supports history, undo, layer-aware actions, and transcript clearing.
• Stand up a new Azure AI Agent Service backend: shared provider wiring, REST routes, setup helpers, and JSON tool registry with validation.
• Package Agent tool metadata (layer mean/diff, contour visualization, layer toggles, etc.) so the assistant can invoke MMGIS actions through a registry-driven dispatcher.
• Update Docker and local runtime to support Azure CLI authentication, mount token caches, and expose agent-specific configuration.
• Refresh project docs/assets (tool configs, sample env entries, PDF reference) to surface the new Copilot capability.

Key Changes

• src/essence/Tools/AgentChat/: UI implementation, renderer utilities, config, and tests covering renderer selection.
• API/Backend/Agent/: new Azure integration (azureService.js), provider abstraction, Express routes, setup bootstrap, and JSON tool specs.
• Tool registry automation and validation via scripts/generate-tool-registry.js, scripts/test-provider.js, scripts/test-ajv.js, and scripts/test-e2e-agent.js.
• Infrastructure: Dockerfile installs Azure CLI and mounts credentials; docker-compose.yml provisions volumes for token caching; sample.env documents required agent variables.
• Dependencies: introduce @azure/* SDKs (AI Agents, Identity, Core Auth), agent tooling libraries, and the new UI dependency echarts.

Describe the feature request

I need or want [...]