Run Gemini transcription with Sapat in Daytona

[xhh678876]

/claim #13

Summary

• Adds a 1,900+ word Daytona guide for running a Gemini-backed Sapat transcription workflow in a reproducible workspace.
• Adds a companion definition for Gemini audio transcription.
• Adds first-time author metadata and original SVG assets for the workflow diagram and author avatar.
• Links the companion Sapat provider implementation: feat: add Gemini transcription provider nibzard/sapat#35

Why this is distinct for #13

I checked the current issue comments and Sapat/content PRs before starting. I did not find an existing Gemini provider slice. This guide focuses on Gemini's multimodal generateContent flow, inline audio size limits, prompt hints for engineering recordings, and validation without spending API credits.

Validation

• npx markdownlint guides/20260521_guide_run_gemini_transcription_with_sapat_in_daytona.md definitions/20260521_definition_gemini_audio_transcription.md authors/haohui_xie.md
• git diff --check

Companion Sapat validation

• python -m venv .venv && .venv/bin/python -m pip install -q -e .
• .venv/bin/python -m unittest discover -s tests -v (12 tests)
• .venv/bin/python -m compileall src tests
• git diff --check

AI-assisted with Codex and Claude Code; I reviewed the article, provider diff, and validation locally before submitting.