Allow users to select Audio Device

[tyrope]

(Note: Any time I refer to "VA" I mean VoiceAttack)

What happens now

[Describe what happens now and any problems that it causes]
Currently, EDDI always uses Windows' Default Output Device.
While I'm streaming Elite to my friends in Discord, I have Voice Attack and my mic loop into Voicemeeter so they can hear VA as well as the game; unfortunately, despite being installed as a VA plugin, EDDI doesn't use it's settings.. So while my friends CAN hear any speech from VA directly, they cannot hear EDDI.

What I'd like to happen

[Describe what you would like to happen. Feel free to include screenshots, videos, markups, alternative features, etc]
Option A[preferred]) A drop-down to implement which Audio Device is selected could perhaps be put in the "Text-to-Speech" config page, between these two settings.

Option B) If installed as a VA plugin, output anything that should be spoken to VA's engine, so all TTS is coming from one program.

How it can happen

[Describe how we could implement the feature, if known. Feel free to include any alternative implementation methods you've considered]
I've peeked into the code a little bit... No idea where to start. :/

EDDI Version

[Please give the full version number here, not just "latest" or "current"]
EDDI v4.0.1
VoiceAttack 1.10.3 64-bit

[Tkael]

Option A could be complicated to implement. We could take inspiration from https://github.com/EDDiscovery/BaseUtilities/tree/5125967c231f4d319e5bf7f28f97d662a15a4359/Audio (which takes a very different approach from EDDI but uses many of the same libraries).

Option B does not appear to be possible as we do not seem to have the ability to either send audio to VoiceAttack or extract the current selected audio device from VoiceAttack settings.