Improve clarity of mouse routing commands

[LeonarddeR]

See also #8479

Steps to reproduce:

Execute the move mouse to navigator object command (laptop layout: nvda+shift+m)

Actual behavior:

The mouse is moved to the review position.

Expected behavior:

According to the doc string, you'd expect that the mouse is moved to a random or start position of the navigator object

Additional thoughts

1. As the review position and navigator object are intertwined, it might make sense to change the doc string of the move mouse to navigator object to make it mention that if possible, the exact position of the review cursor within the navigator object is targetted.
2. In line with this, it might be an idea to change the move navigator object to mouse script in such a way that it aims to position the review cursor at the exact position of the mouse (i.e. by creating a TextInfo at the mouse position). This would also open the possibility to position the mouse and than change the screen review position to the position of the mouse, for example.

[LeonarddeR]

cc @josephsl, @feerrenrut and @Qchristensen, your thoughts are especially welcome here.