magnitudedev/magnitude

AI Summary: Implement a feature in the Magnitude test runner to display the AI agent's thought processes during test execution, similar to the existing 'narrate' functionality in the `startBrowserAgent` method. This will allow developers to debug and understand the reasoning behind the agent's actions.

AI Summary: Implement functionality to export test results in JSON and HTML formats. The reports should include screenshots and a history of agent actions alongside the test results.

AI Summary: Implement beforeEach, beforeAll, afterEach, and afterAll hooks in the Magnitude testing framework, mirroring the functionality found in other test runners like Jest or Mocha. This will allow users to execute setup and teardown code before and after individual tests or the entire test suite.

AI Summary: Refactor the Magnitude test runner to improve test isolation. This involves replacing the current esbuild-based compilation and execution method with a more robust approach that runs tests in separate processes, potentially leveraging techniques similar to Ava.js. This change will enhance test reliability and allow for the reintroduction of the `ink` CLI rendering library.