fix: many miscellaneous fixes to integrations and agents API

[creatorrr]

[!IMPORTANT] Refines Remote Browser functionality by removing wait_for_load, updating models, and enhancing PlaywrightActions with error handling and screenshot capabilities.

• Behavior:
  • Removed wait_for_load action from RemoteBrowserArguments and RemoteBrowserArgumentsUpdate.
  • Updated task_to_spec() in tasks.py to exclude task_id, id, and agent_id.
  • Modified create_task() and create_or_update_task() to return query results directly.
• Models:
  • Updated RemoteBrowserSetup to remove RemoteBrowserSetupUpdate and its fields.
  • Refined RemoteBrowserOutput to include output, error, base64_image, and system fields.
• Functions:
  • Enhanced PlaywrightActions with _with_error_and_screenshot decorator for error handling and screenshots.
  • Added methods _get_screen_size(), _set_screen_size(), _wait_for_load(), _execute_javascript(), _set_window_vars(), _get_mouse_coordinates(), and _get_element_coordinates().
• Misc:
  • Removed spec_to_task import from get_task.py and list_tasks.py.

^{This description was created by for c8ef19679ce4ed096486264ccbf21f0b8d59b41b. It will automatically update as commits are pushed.}

[sweep-ai[bot]]

Hey @creatorrr, here is an example of how you can ask me to improve this pull request:

@sweep Add unit tests for the `_with_error_and_screenshot` decorator in `PlaywrightActions` class to verify:
- Error handling and proper error message propagation
- Screenshot capture during successful operations
- Proper combination of output and screenshot in the returned RemoteBrowserOutput

:book: For more information on how to use Sweep, please read our documentation.