Need to handle "quite" when it's a typo for "quit"

This is a typo I've caught myself making and I've noticed Google Chrome's built-in grammar checker flags it at least some of the time.

This might be integrated into the linter than checker for "quite" vs "quiet" typos, or might be better as a standalone linter.

Examples found on GitHub:
- [BUG] close window != **quite** app
- Error no valid update available, can't **quite** and install
- Quit Electron to continue updating? [Not Now] [**Quite** and Update]. Clicking either button closes the popup but it comes back in a minute or so.
- No visible close or **quite** command
- ... if you edit the AGENTS.md file you need to **quite** and restart copilot cli.
- I tried to **quite** and got this:
- **Quite** and re-open VSCode.
- Had to **quite** and reopen to fix.
- If I **quite** and re-open alt-tab it will correctly find the meeting window and ...
- As the tittle, can rasa-core interactive learning export current conversations as stories and continue. Now everytime it **quite** and I have to retrain the dialogue. Tks!
- even when Stop button is not working, if I **quite** the app by closing it's window
- And I have the **quite** the terminal app to stop it.
- when I **quite** the app and it becomes a background process, but the momery is still to high
- When notepad++ shutdown abnormal (application crash or Windows shutdown force **quite** the application), last session (opened tabs etc) will lost.
- The only way to exit vim is to close the tmux session/window or to **quite** the whole terminal
- Please advise a proper way to implement the setting, so that accidental won't **quite** the entire shell.
- I can force **quite** the app, to get the messages to load after quitting a few times, but it does not load right away until I quit.

Start by locating the existing checker or linter rule for “quite” versus “quiet” and the tests covering typo corrections. Compare its design with the requested “quite” for “quit” cases and determine whether the behavior belongs there or in a standalone rule. Done means the reported examples are detected with appropriate corrections and regression tests cover them.

Written by the indexing model from the issue text.