feat: real streaming asr fixes

[pixelsama]

Summary

• keep output handler connection alive until streaming audio completes
• add asyncio import to dialog-engine app to allow scheduling background TTS tasks
• document the TTS audio debug plan so others can reproduce the test flow

Testing

• python script replayed demo/sauc_python/input.webm through ws/input and ws/output: received 500+ audio chunks and audio_complete
• frontend mic capture verified text stream and audio chunk metadata in console

[pixelsama]

@codex review the code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Fix invalid numpy version specifier

The new requirements entry numpy>=2.0.2<3 is not valid pip syntax because range specifiers must be separated by a comma (numpy>=2.0.2,<3). As written, pip install -r requirements.txt will fail to parse the requirement and the dialog-engine service cannot build. This blocks any environment setup that relies on the file.

Useful? React with 👍 / 👎.