Add an Ollama local provider adapter #39

Goal

Support a lighter local-provider path in addition to LM Studio.

Depends on

• Define a provider-neutral AI client and capability contract #35, Move AI execution and credentials into a secure Electron main-process service #36, Add provider profiles and connection controls to Settings #37, and Migrate LM Studio to the provider abstraction without behavior regressions #38.

Scope

• Connect to a loopback Ollama server.
• Discover installed models and test connectivity.
• Support streaming and cancellation.
• Declare structured-output and usage capabilities honestly for the detected server behavior.
• Provide actionable setup and model-not-found errors.

Acceptance criteria

• A user can select Ollama, choose an installed model, test it, and run all three Lucid lanes.
• Command Mode is disabled or clearly warned when the selected configuration cannot meet its structured-output safety requirements.
• No automatic fallback to LM Studio or a cloud provider occurs.