Add self-describing key input help

[NSoiffer]

NVDA has a mode where pushing a key says what the key does. The implementation doesn't support that, but it shouldn't be hard to implement. Obviously, the strings need to be translatable ones.

[NV-Codes]

I have not worked with MathCAT's or NVDA's code before, but I had the following idea:

1. Make a file addon/globalPlugins/MathCAT/navCommands.py that defines a dictionary with command-description key-value pairs.
2. In MathCAT.py, import the dictionary and create a conditional block that is executed if a) NVDA's help mode is active and b) MathCAT's navigation mode (for an expression) is active.
   • If a key for which MathCAT has no binding is pressed, pass the gesture to NVDA.
   • Otherwise, have the screen reader say and braille the description (dictionary value) that is mapped to any key pressed (dictionary key).

Dictionary Pseudo-Code:

commandsDictionary = {
'space': 'Speaks the current navigation node',
'control C': 'Copies the math content of the current node to the clipboard'
}

MathCAT Pseudo-Code:

if NVDA_HelpModeIsActive and MathCAT_NavigationModeIsActive:
    if inputGesture not in commandsDictionary: pass_to_NVDA
    speak(commandsDictionary[inputGesture])

I do not know how to query

• whether NVDA's help mode is active.
• whether MathCAT's navigation mode is active.
• the user's input gestures.

I would also need to know how to pass an input gesture to NVDA and how to pass a string to be spoken and brailled by the screen reader, though if these operations are straightforward (and if my idea is feasible), then perhaps I could try implementing this feature as a "good first issue."