feat: editorial review dashboard + elite-grade pilot batch (5 SKUs)

Ships the second dashboard surface — a Pattern Library + Preview Theatre — that
presents the 4-section PDP pilot batch back to Umar, compliance, and the board
in an editorial format. Adds the full data layer that drives it: 5 source-backed
per-SKU drafts at QA 100/100, 15 competitor PDP semantic extracts, PubMed
evidence packs, EFSA claims library extension, JV brand voice guide, hand-curated
product FAQs, and the Matrixify-ready CSV exports for Lewis.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

content_population_exports/dataforseo_pilot_plan.md

@@ -0,0 +1,26 @@
+# DataForSEO Stage 1 pilot packet
+
+Generated: 2026-05-20T16:41:07.180Z
+
+This packet prepares the first paid source acquisition stage without spending credits. It does not call DataForSEO.
+
+## Guardrails
+- Stage 1 target cap: 12
+- Recommended first depth: 100 reviews per ASIN
+- Stop after the first payload/run and inspect raw JSON quality + cost before increasing toward 1000 reviews per ASIN.
+- Store credentials only in shell environment variables; do not commit credentials.
+
+## Current status
+- Stage 1 targets: 10
+- Ready task payload rows: 6
+- Missing ASINs: 4
+
+## Files
+- `data/sources/dataforseo/asin-map.csv` - fill ASIN/ProductUrl before task creation.
+- `content_population_exports/dataforseo_amazon_reviews_task_payload.json` - generated DataForSEO task payload; empty until ASINs are filled.
+- `content_population_exports/dataforseo_amazon_reviews_curl_template.sh` - safe curl template using env credentials.
+- `data/sources/dataforseo/amazon-review-output-template.json` - local raw-output shape expected by source validators.
+
+## DataForSEO notes
+- Amazon reviews task creation requires an ASIN plus location and language parameters.
+- Keep `depth` between 100 and 1000 for top-3 research runs; increase only after output quality is proven.