Is your feature request related to a problem? Please describe.

Recently, Unicode normalization has been added to NVDA, but in #16466, there was some discussion about whether it should be enabled by default.
Furthermore, there was some debate about character navigation, i.e. whether normalization should be enabled by default for character navigation, and/or whether the delayed character description feature could be utilized to some extend.

Describe the solution you'd like

1. Enable normalization by default for speech output.
2. When enabled, always normalize (i.e. also with character navigation) but prepend the word 'normalized' or something less technical to the character.

Describe alternatives you've considered

@CyrilleB79 proposed a three choice option in #16466 (comment):

1. Normalization disabled (Current default)
2. Normalization enabled to read text but disabled to read by character (current behavior of enabled)
3. Normalization enabled for all (not yet implemented)

I'm happy to consider this alternative as well, though it will probably take more effort. I'm now at the point where I think if we add a proper term prepended for character navigation, that would be the best solution.

Additional context

1. I'm reluctant to take the character descriptions feature into account:

   • Current use of these is for phonetics. I'm inclined to say that adding Unicode character descriptions is a fundamental change to this feature
   • There are no translated names for the several mathematical Unicode chars in CLDR

2. Braille is beyond discussion.

Discussion guidelines

When attending to the discussion, please avoid repeating standpoints over again that were already stated in #16466. It is clear that we have different opinions and that is okay.