chore: remove upload instruction, video already on YouTube

chore: set YouTube video ID jgYXyij1P9Q

chore: update final video (44s, 720p30, audio)

chore: update video ref to manim version (33s, 720p)

docs: add @ilhamfputra credit for pasal.id API on Kredit tab

docs: add YouTube guide video section to Beranda tab

feat: add ethics guardrails, verifier, beranda + kredit tabs

feat: add token setup guide in Settings tab for new users

fix: don't embed HF_TOKEN/PASAL_API_TOKEN in frontend HTML/JS

fix: shared global rate limiter for pasal.id across all instances (1.5s delay)

fix: enable Gradio queue for progress reporting

fix: add rate limiter + retry on 429 to pasal.id client

fix: detect model IDs (no ://) as HF InferenceClient, not OpenAI

fix: add Settings import in _pasal_settings

fix: add missing LLMConfig/Settings import in custom endpoint branch

SMALL→Qwen3.5-9B, add extra_body(enable_thinking=False) to HFLLM

Fix: use huggingface_hub InferenceClient as default backend

Switch to Qwen3-32B (BIG) and Qwen3-8B (SMALL) — confirmed working on HF API

Fix load_ruu_text: guard against long text being treated as file path

Fix build_pool: don't call load_settings(), use env vars/HF defaults directly

Swap defaults: BIG→Qwen3.6-27B, SMALL→Qwen3.6-35B-A3B

Remove -e . from requirements, add rich dep

Drop sdk_version — let HF resolve from requirements.txt

Fix colorTo value for HF metadata

Bump Gradio to 6.x (latest is 6.14)

Add HF Space for Build Small Hackathon

Expand canonical probes to remaining 10 sectors + fix surat strict toggle

Add trusted_recent override for fresh regulations not yet on pasal.id

Test 6: MBG Keracunan Grobogan — multi-Komisi + sosial sector validation

Add corpus-drift regression check script + baseline

Inject DPR Komisi 2024-2029 portfolio reference into agent prompts

Test 5: Bekasi train+EV taxi accident — transport sector + extractor fixes

Add amendment-chain probe to citation verifier (advisory only)

Extend canonical FRBR-URI probes to dosen and medical-education sectors

Add Test 4 — PPDS dan internsip dokter (cross-sector kesehatan validation)

Add Test 3 — gaji dan tunjangan dosen (PT-domain validation)

Correct Kasus Ibam case status: tuntutan, not vonis

Tighten verifier: institutional-context tokens don't count as topical match

Add SMALL-model domain classifier for Penyusun

Add Kasus Ibam whitepaper — failure analysis, hardening, post-fix validation

Harden citation pipeline + add comprehensive test suite

Add whitepaper documenting first live validation run

Fix three Windows/Qwen3 issues uncovered by first live run

Add roadmap entry for Pemantau Berita (5th agent)

Initial commit: legawa multi-agent system for Indonesian legislators