Simplify the Text vs Voice split and auth

[renal128]

Summary

Voice goes through WebRTC, Text goes through Websocket.
Auth is transport-dependent - WebRTC and Websocket have different available auth methods (auth doc).
So the API was a bit awkward and the auth <> conversation type relationship wasn't properly enforced.

Now the API is cleaner:

• we have ConversationAuth.Voice and ConversationAuth.TextOnly instead of a single ConversationCredential that had all available methods from both transport mixed in one enum
• we have public startVoiceConversation(auth: ConversationAuth.Voice, ...) and startTextOnlyConversation(auth: ConversationAuth.TextOnly, ...) methods that enforce the modality and auth match

Test plan

• swift test
• swift test -Xswiftc -swift-version -Xswiftc 6
• swiftformat . --strict
• Widget modes using the shared auth types

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit a72eef5. Configure here.}