.Net: ADR for audio abstraction and implementation renaming

[dmytrostruk]

Motivation and Context

Added ADR for audio abstraction and implementation renaming for further discussion and agreement.

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the SK Contribution Guidelines and the pre-submission formatting script raises no violations
• All unit tests pass, and I have added new tests where possible
• I didn't break anyone 😄

[Krzysztof318]

I would add content about live/continuous streaming audio-to-text and text-to-audio, how it should be abstracted, to this ADR.
This PR #5094 proposed a new class AudioStreamContent.
Abstraction should support full real time streaming, when we sending an audio stream and service returns IAsyncEnumerable<StreamingTextContent>

[dmytrostruk]

I would add content about live/continuous streaming audio-to-text and text-to-audio, how it should be abstracted, to this ADR. This PR #5094 proposed a new class AudioStreamContent. Abstraction should support full real time streaming, when we sending an audio stream and service returns IAsyncEnumerable<StreamingTextContent>

I believe it's out-of-scope of this ADR, because currently we want to align our vision in terms of naming. Regarding streaming, my initial thoughts are to implement it in similar way how it's done in IChatCompletionService at the moment. One interface method for non-streaming scenario and one for streaming scenario. Method signature should be similar to other abstractions we have at the moment. But it's just general ideas, in order to understand better how streaming abstraction should look like, we have to take a couple of AI providers which support that functionality, compare their API shape and try to find common denominator. This work may potentially take separate ADR if needed and I'm not sure what priority it has at the moment.