What I can still say about the error message in general is that when you see “Start request repeated too quickly.” it means the systemd service already restarted so many times you don’t even see the original error message since the current issue is that it can’t start again yet. You should scroll up as much as you can in the journalctl output or use the --since option and possibly the --lines as well to get older logs.
What I saw usually is that when the “repeated too quickly” error message appeared, there was a syntax error in the daemon json, but unless your daemon.json is in the “storage folder”, that is probably not what happened in your case.
If you don’t find useful error messages, stop the docker daemon instead of restarting and run the command defined in the systemd service manually. Or if you are not sure what to run, wait enough so the systemd service can start again, start Docker and when it fails, it will probably not restart. But to be safe you can run systemctl stop docker docker.socket as well.