Add custom endpoint support for OpenAI-compatible APIs

[Copilot]

This PR adds support for custom API endpoints to the OpenAI provider, allowing users to connect to OpenAI-compatible APIs such as Azure OpenAI, local LLM servers (Ollama, LM Studio), or other hosted services.

Changes Made

Configuration Enhancement:

• Added base_url field to the ProviderConfig struct
• Extended config management with GetBaseURL() and SetBaseURL() functions
• Updated YAML configuration to support the new base_url parameter

Provider Updates:

• Enhanced OpenAIProvider with NewOpenAIProviderWithBaseURL() function
• Maintains backward compatibility with existing NewOpenAIProvider() function
• Custom base URL is passed through to the OpenAI client when specified

CLI Improvements:

• Interactive config (lazycommit config set) now prompts for custom base URL when using OpenAI provider
• lazycommit config get displays the configured base URL when set
• Fixed missing openrouter case in the commit command

Documentation:

• Updated README with configuration examples for Azure OpenAI and local servers
• Added dedicated "Custom Endpoints" section with practical examples
• Clarified that OpenAI provider supports custom endpoints via base_url

Usage Examples

Azure OpenAI:

providers:
  openai:
    api_key: "$AZURE_OPENAI_API_KEY"
    model: "gpt-4"
    base_url: "https://your-resource.openai.azure.com/openai/deployments/your-deployment/chat/completions"

Local LLM Server:

providers:
  openai:
    api_key: "dummy-key"
    model: "llama2"
    base_url: "http://localhost:11434/v1"

The feature maintains full backward compatibility - existing configurations continue to work unchanged, and the base URL is optional.

Fixes #17.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.