Now allows for changing the model and voice in proxy mode.

Improved playback state feedback for proxy server audio.

Added a new option to utilize a proxy server. This enables near real time TTS playback without having to wait for your entire message to be processed.
This feature is enabled with the use of an openedai-speech-proxy server. Detailed setup instructions are available at the repository page. https://gitlab.sfnet.app/dstewart/openedai-speech-proxy

Addressed a bug where overlapping container IDs would result in undesirable behavior on certain websites.

- Added greater Visual feedback for TTS Playback.
- TTS playback is now stoppable from the popup menu.