feat: add streaming STT service with Whisper backend

- stt_streaming.py: HTTP-based STT using external Whisper service
- stt_streaming_local.py: ROCm-based local Whisper inference
- Voice Activity Detection (VAD) with WebRTC
- Interrupt detection for barge-in support
- Session state management (listening/responding)
- OpenTelemetry instrumentation with HyperDX support
- Dockerfile variants for HTTP and ROCm deployments

requirements-rocm.txt

@@ -0,0 +1,24 @@
+# Core dependencies
+nats-py>=2.0.0,<3.0.0
+msgpack
+
+# Whisper for local STT inference (uses PyTorch already in base image)
+openai-whisper>=20231117
+
+# Audio processing
+soundfile
+numpy
+
+# OpenTelemetry core
+opentelemetry-api
+opentelemetry-sdk
+opentelemetry-exporter-otlp-proto-grpc
+opentelemetry-exporter-otlp-proto-http
+opentelemetry-instrumentation-logging
+
+# HyperDX support (uses OTLP protocol)
+# HyperDX is compatible with standard OTEL exporters, just needs API key header
+opentelemetry-sdk-extension-aws  # For additional context propagation
+
+# HTTP health server for kserve compatibility
+aiohttp