feat(uno-q): bundle the granular skill set alongside arduino-uno-q on every node

Vendors the fork's 11 granular UNO Q skills (bridge, flashing, led-matrix,
uno-q-hardware, sketch-patterns, modulino, linux-led, audio, vision, wireless,
arduino-app-lab) next to the comprehensive arduino-uno-q skill, and installs the
whole set into every agent's workspace on each board.

Why both: the comprehensive skill is the rich cloud reference (read_skill →
references); the granular skills are keyword-triggered and match the fork's eager
skill-inliner rules, so the on-board Qwen auto-inlines them (no read_skill
round-trip). flashing + led-matrix carry the exact uno_q_flash + frame-API /
ArduinoGraphics-not-installed detail that makes flashing reliable.

- push-skill.sh generalized: a single skill dir (has SKILL.md) OR a parent dir
  installs every skill under it; provision-fleet now ships all of skills/.
- Verified on board 65301572: cloud/Sonnet-5 lists all 12 skills.

Co-Authored-By: Claude Opus 4.8 <[email protected]>

deploy/uno-q/skills/audio/SKILL.md

@@ -0,0 +1,31 @@
+---
+name: audio
+description: Audio on the Arduino Uno Q — microphone capture and audio output via ALSA on the Linux side, plus speech (whisper/vosk/espeak). Load this for recording sound, playing audio, or speech-to-text / text-to-speech on the board.
+---
+
+# Audio on the Uno Q (Linux side)
+
+Audio endpoints (mic, headphone/line-out via JMISC) are **Linux/MPU** devices,
+driven through **ALSA**. Sketches on the MCU do not handle audio.
+
+## Record / play (ALSA)
+
+```sh
+arecord -d 5 -f cd /tmp/clip.wav          # record 5s
+aplay /tmp/clip.wav                        # play
+arecord -l                                 # list capture devices
+```
+
+## Speech (Python, MPU)
+
+- **STT:** `whisper` (or `vosk` for a lighter offline model).
+- **TTS:** `espeak` / `espeak-ng` (offline, no key).
+
+```sh
+espeak-ng "Sketch flashed."
+```
+
+## Pitfalls
+
+- No Arduino audio libraries — audio is a Linux-side capability.
+- Keep models small; the MPU is also hosting the LLM (~1.1 GB RSS) on the 4 GB SKU.