feat: M2 zero-copy foundation — EGL→CUDA import + NVENC CUDA-frame path

Scaffolding for dmabuf zero-copy (plan §9), opt-in via LUMEN_ZEROCOPY:

- src/zerocopy/{cuda,egl}.rs: hand-rolled CUDA Driver-API FFI (no Rust crate
  exposes the EGL-interop calls / CUeglFrame) with a shared process-wide
  CUcontext + pitched device buffers; an EGL importer (GBM platform on the
  NVIDIA render node) that turns a dmabuf into an EGLImage, registers it with
  CUDA, and copies it device-to-device into an owned buffer. `zerocopy-probe`
  subcommand validates the FFI/linking/GPU access — confirmed on the box
  (driver 595, EGL_EXT_image_dma_buf_import + modifiers).
- CapturedFrame gains a FramePayload enum (Cpu(Vec<u8>) | Cuda(DeviceBuffer));
  the encoder branches: CPU keeps the expand+upload path, CUDA wraps the device
  buffer in an AV_PIX_FMT_CUDA frame fed straight to hevc_nvenc (sharing our
  CUcontext via a hand-declared AVCUDADeviceContext, since ffmpeg-sys doesn't
  bind hwcontext_cuda.h). open_video/the encoder take a `cuda` flag derived from
  the first frame's payload.

The capture-side dmabuf negotiation (which produces the Cuda frames) is the
next step; the CPU path is unchanged and remains the default + fallback. Builds
clean, clippy clean, tests pass.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Cargo.lock

@@ -1356,6 +1356,16 @@ dependencies = [
  "wasm-bindgen",
 ]
 
+[[package]]
+name = "khronos-egl"
+version = "6.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6aae1df220ece3c0ada96b8153459b67eebe9ae9212258bb0134ae60416fdf76"
+dependencies = [
+ "libc",
+ "libloading",
+]
+
 [[package]]
 name = "latency-probe"
 version = "0.0.1"
@@ -1501,6 +1511,7 @@ dependencies = [
  "ffmpeg-next",
  "futures-util",
  "hex",
+ "khronos-egl",
  "libc",
  "lumen-core",
  "mdns-sd",