Add AI workflow md2video demo generator

[leether]

Summary

• add a markdown script for an AI workflow short video demo
• add a reproducible generator that checks env presence, prefers Dreamina/Jimeng fast VIP, and falls back to local scenes when external generation fails
• keep generated video and evidence under ignored output paths

Verification

• .venv/bin/python -m py_compile scripts/generate_ai_workflow_video.py
• .venv/bin/python scripts/generate_ai_workflow_video.py --input examples/ai_workflow_video_script.md --output-dir output/ai_workflow_demo --dreamina-model seedance2.0_vipfast --dreamina-submit-poll 90 --dreamina-result-wait 45 --max-dreamina-scenes 4
• .venv/bin/python scripts/generate_ai_workflow_video.py --input examples/ai_workflow_video_script.md --output-dir output/ai_workflow_demo --dreamina-model seedance2.0_vipfast --force-local
• .venv/bin/python scripts/preflight.py --input examples/ai_workflow_video_script.md --skip-command-checks --json
• .venv/bin/python scripts/smoke_imports.py
• bash scripts/privacy-check.sh
• ffprobe -v error -show_entries format=duration:stream=index,codec_type,codec_name,width,height,r_frame_rate,duration -of json output/ai_workflow_demo/ai_workflow_demo.mp4

Local Artifact

• Video: /Users/lize/workspace/md2video/output/ai_workflow_demo/ai_workflow_demo.mp4
• Manifest: /Users/lize/workspace/md2video/output/ai_workflow_demo/run_manifest.json
• Compliance: /Users/lize/workspace/md2video/output/ai_workflow_demo/compliance_report.json