feat(p1): scaffolding + Device Manager / VRAM probe + CLI detect

- pyproject (uv, src layout) + extras: engine/gpu/api/diarize/llm
- config.py (pydantic-settings, SCRIBE_ env)
- devices/: vram_probe (NVML/psutil/disk) + DeviceManager →
  capability tier T0–T3, precision by cc/VRAM, worker estimate (계획 §3.6, AC-2/3)
- cli.py (typer): detect (구현) + transcribe/bench/serve (스텁)
- run.sh, .env.example, README

Verified on GTX 1050/2GB: detect → T0_CPU (turbo doesn't fit → explicit
downgrade, fail-explicit). Overrides (--device/--workers) work. 7 unit tests
cover T0–T3 + overrides via synthetic VRAM. ruff clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

pyproject.toml

@@ -0,0 +1,38 @@
+[project]
+name = "luke-scribe"
+version = "0.1.0"
+description = "내부용 로컬 STT 전사 API (faster-whisper, hardware-adaptive)"
+requires-python = ">=3.11"
+dependencies = [
+    "pydantic>=2.7",
+    "pydantic-settings>=2.3",
+    "typer>=0.12",
+    "rich>=13.7",
+    "psutil>=5.9",
+    "nvidia-ml-py>=12.535",
+    "huggingface-hub>=0.24",
+]
+
+[project.optional-dependencies]
+# 엔진 — transcribe/bench 증분에서 설치 (uv sync --extra engine)
+engine = ["faster-whisper>=1.0.3", "av>=11"]
+# GPU CUDA 런타임 (faster-whisper GPU 추론 시)
+gpu = ["nvidia-cublas-cu12", "nvidia-cudnn-cu12"]
+# P2 API + Queue
+api = ["fastapi>=0.110", "uvicorn[standard]>=0.29", "redis>=5.0", "rq>=1.16"]
+# P5 옵션
+diarize = ["pyannote.audio>=3.1"]
+llm = ["openai>=1.30"]
+
+[project.scripts]
+luke-scribe = "luke_scribe.cli:main"
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/luke_scribe"]
+
+[dependency-groups]
+dev = ["pytest>=8.2", "ruff>=0.5"]