docs(host): prove unsafe blocks in the Windows + cross-platform files + gate them (unsafe-proof program 3/N)

Continues the unsafe-proof program across the Windows/cross-platform host files
(~75 blocks, 21 files), each with a SAFETY proof of the real invariant and a
per-file #![deny(clippy::undocumented_unsafe_blocks)] gate:

  capture/windows: dxgi.rs, wgc_relay.rs, wgc.rs, desktop_watch.rs, composed_flip.rs
                   (windows-rs COM: interface validity, same-D3D11-device textures,
                    immediate-context single-thread, borrowed args outlive the call)
  windows: service.rs (SCM/token/CreateProcessAsUserW/event handles — OwnedHandle
           liveness, no double-close/signal race), win_display, wgc_helper, interactive
  vdisplay/windows: manager.rs, pf_vdisplay.rs (SwDeviceCreate/IddCx/ioctl handle
                    liveness via the OnceLock VDM singleton + OwnedHandle)
  encode/windows: ffmpeg_win.rs (full AVBufferRef refcount audit — balanced, NO leaks,
                  unlike the vaapi sibling), sw.rs
  cross-platform: gamestream/audio.rs (libopus), gamestream/stream.rs (sendmmsg),
                  inject/windows/sendinput.rs, audio/windows/wasapi_mic.rs,
                  session_tuning.rs, vdisplay.rs

Two findings (handled separately):
- wgc_relay.rs `unsafe impl Sync for HelperRelay` is UNSOUND (its mpsc Receiver is
  !Sync) though not live-exploited — marked SUSPECT inline; fix pending box check
  (it touches the in-flight punktfunk1.rs).
- capture.rs / encode.rs (PARENT modules of the WIP idd_push.rs / nvenc.rs) do NOT
  get the file deny yet — it would propagate the lint into the undocumented WIP
  children. The deny lands there once those are documented (after the WIP commits).

Linux-visible parts verified green (cargo clippy -p punktfunk-host --all-targets
-- -D warnings). The cfg(windows) deny gates are box-verified next.

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

crates/punktfunk-host/src/encode/windows/sw.rs

@@ -2,6 +2,8 @@
 //! fallback when NVENC is unavailable). Low-latency screen-content config: single-reference,
 //! no B-frames (Baseline), bitrate rate-control, in-band SPS/PPS each IDR, BT.709 limited range.
 //! Synchronous: `submit` encodes immediately and stashes the AU for `poll` (no internal queue).
+// Every `unsafe` block in this file carries a `// SAFETY:` proof; enforce it (unsafe-proof program).
+#![deny(clippy::undocumented_unsafe_blocks)]
 
 use super::{EncodedFrame, Encoder};
 use crate::capture::{CapturedFrame, FramePayload, PixelFormat};
@@ -30,6 +32,12 @@ pub struct OpenH264Encoder {
 }
 
 // openh264's Encoder holds a raw C handle (not auto-Send); it lives on the single encode thread.
+// SAFETY: `OpenH264Encoder` wraps `Oh264` (openh264's `Encoder`), which holds a raw C handle to the
+// openh264 `ISVCEncoder` and is not auto-`Send`; the other fields (`YUVBuffer`, `Vec`, scalars,
+// `Option<EncodedFrame>`) are plain owned data. The session creates the encoder, calls
+// `submit`/`poll`/`flush`, and drops it all on one dedicated encode thread, never sharing it by
+// reference across threads, so the C handle is only ever touched from a single thread. Moving the
+// whole value to that thread is therefore sound — there is no concurrent access to the handle.
 unsafe impl Send for OpenH264Encoder {}
 
 impl OpenH264Encoder {