feat: Add support for default model per provider

[crazywolf132]

Description

This PR implements the feature requested in issue #1031 to allow setting default models per provider.

Changes

• Added providerDefaultModels field to the configuration to store default models per provider
• Added defaultModel property to each provider in the providers config
• Modified the CLI code to use the default model for the selected provider when switching providers
• Updated the documentation in README.md to explain the new feature
• Added comprehensive tests to verify the functionality

How It Works

Now you can configure default models for each provider in two ways:

1. Using the providerDefaultModels object in your config:

   {
     "providerDefaultModels": {
       "openai": "o4-mini",
       "ollama": "qwen2.5-coder:14b",
       "mistral": "mistral-large-latest"
     }
   }

2. Or by adding a defaultModel property to each provider:

    {
      "providers": {
        "openai": {
          "name": "OpenAI",
          "baseURL": "https://api.openai.com/v1",
          "envKey": "OPENAI_API_KEY",
          "defaultModel": "o4-mini"
        }
      }
    }

When you switch providers using the /model command or the --provider flag, Codex will automatically use the default model for that provider if one is configured.

Additionally, when you select a model for a provider, it will be saved as the default model for that provider for future use.

Fixes #1031