Initial implementation of AudioCraft Studio

Complete web interface for Meta's AudioCraft AI audio generation:

- Gradio UI with tabs for all 5 model families (MusicGen, AudioGen,
  MAGNeT, MusicGen Style, JASCO)
- REST API with FastAPI, OpenAPI docs, and API key auth
- VRAM management with ComfyUI coexistence support
- SQLite database for project/generation history
- Batch processing queue for async generation
- Docker deployment optimized for RunPod with RTX 4090

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

.env.example

@@ -0,0 +1,42 @@
+# AudioCraft Studio Configuration
+# Copy this file to .env and customize as needed
+
+# Server Configuration
+AUDIOCRAFT_HOST=0.0.0.0
+AUDIOCRAFT_GRADIO_PORT=7860
+AUDIOCRAFT_API_PORT=8000
+
+# Paths (relative to project root)
+AUDIOCRAFT_DATA_DIR=./data
+AUDIOCRAFT_OUTPUT_DIR=./outputs
+AUDIOCRAFT_CACHE_DIR=./cache
+
+# VRAM Management
+# Reserve this much VRAM for ComfyUI (GB)
+AUDIOCRAFT_COMFYUI_RESERVE_GB=10
+# Safety buffer to prevent OOM (GB)
+AUDIOCRAFT_SAFETY_BUFFER_GB=1
+# Unload idle models after this many minutes
+AUDIOCRAFT_IDLE_UNLOAD_MINUTES=15
+# Maximum number of models to keep loaded
+AUDIOCRAFT_MAX_CACHED_MODELS=2
+
+# API Authentication
+# Generate a secure random key for production
+AUDIOCRAFT_API_KEY=your-secret-api-key-here
+
+# Generation Defaults
+AUDIOCRAFT_DEFAULT_DURATION=10.0
+AUDIOCRAFT_MAX_DURATION=300.0
+AUDIOCRAFT_DEFAULT_BATCH_SIZE=1
+AUDIOCRAFT_MAX_BATCH_SIZE=8
+AUDIOCRAFT_MAX_QUEUE_SIZE=100
+
+# Database
+AUDIOCRAFT_DATABASE_URL=sqlite+aiosqlite:///./data/audiocraft.db
+
+# Logging
+AUDIOCRAFT_LOG_LEVEL=INFO
+
+# PyTorch Optimization (recommended)
+PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True