This is a nice idea however the "cost" (time and added complexity) to implement it are much higher than the (small) benefit.
The only way that user code directly interacts with the IDE is through the logs. It is possible to send commands through the logs. This is how LogColor works for example (and there are other cases as well).
However a "clear command" cannot work this way as it will cause the logs to clear in unwanted situations. For example when the IDE connects to the logs and then sees an old clear command.