[FEAT] ESP32-Compatible Node

[NeonDaniel]

Objective

A Neon Node implementation written in C that may be run on ESP32-based hardware.

Initial Implementation Requirements

• Configurable or auto-detected microphone input and speaker output
• Continuous listener loop with wake-word (WW) detection and voice activity detection (VAD)
• Makes calls to a HANA Websocket API

Other Considerations

• If WW/VAD performance is not possible then this may stream microphone input to a remote service that handles those tasks (See Implement streaming audio Websocket neon-hana#30)
• An initial implementation may be completed for a Raspberry Pi Zero or similar if ESP32 performance is unacceptable for WW and VAD

[NeonDaniel]

HomeAssistant's implementation might be a good starting point
https://www.home-assistant.io/voice_control/thirteen-usd-voice-remote/
https://www.home-assistant.io/voice_control/s3_box_voice_assistant/

[NeonDaniel]

#14 implemented a WebSocket connection. A node should implement that, rather than HTTP