docker: production image + Gitea Actions for monthly refresh

Dockerfile: self-contained image with corpus + Chroma + BM25 baked
in. Drawbar's compose pulls + runs without volume mounts. Built from
sources.json (labels schema), PRODUCT_NAME=crop_chem by default,
HYBRID_SEARCH=true (always-on for production quality). RERANK_URL +
OLLAMA_URL get set at compose time.

.gitea/workflows/refresh.yml: monthly cron (1st @ 06:00 UTC) does
full scrape → reindex → image push. Scrapes Bayer (~30 min) +
EPA PPLS row-crop filtered (~7h). Skips reindex+push if no corpus
diff. Tags pushed: :latest, :<sha12>, :corpus-<YYYY.MM.DD>.

.gitea/workflows/image-only.yml: on-demand or auto on code-only
pushes to main (paths: docs_mcp/, rag/, scrape/, requirements.txt,
Dockerfile, sources.json). Reindexes from committed corpus, builds
image, pushes. ~10 min vs ~9h full refresh.

.gitignore: corpus/ now COMMITTED (4,159 labels, 265 MB of .md +
sidecars). Lets image-only.yml rebuild indexes without re-scraping.
chroma/ + bm25/ still gitignored (regenerable binary indexes).

.dockerignore: drops venv, eval results, PLAN/README/CLAUDE.md,
deploy/, .git/ — keeps the image lean. corpus + chroma + bm25
explicitly NOT in dockerignore (those go INTO the image).

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

corpus/epa_ppls/279-3146.md

@@ -0,0 +1,62 @@
+# AUTHORITY 4 F HERBICIDE
+
+- EPA Reg No: **279-3146**
+- Registrant: FMC CORPORATION 
+- Signal word: Caution
+- Active ingredients: Sulfentrazone (39.6%)
+- Label accepted: 2010-02-17
+- Source PDF: https://www3.epa.gov/pesticides/chem_search/ppls/000279-03146-20100217.pdf
+
+---
+UNITED STATES ENVIRONMENTAL PROTECTION AGENCY 
+WASHINGTON, D.C. 20460 
+OFFICE OF 
+PREVENTION, PESTICIDES ~D 
+TOXIC SUBSTANCES 
+Callista O. Chukwunenye 
+FMC Corporation 
+1735 Market St. 
+Philadelphia, PA 19103 
+Dear Or. Chukwunenye: 
+SUBJECT: Change in Status of Registrations / 
+Authority 4F - EPA Registration No. 279-3146 
+Authority 75 OF - EPA Registration No. 279-3148 
+Sulfentrazone Technical- EPA Registration No. 279-3149 
+Authority BJ- - EPA Registration No. 279-3175 
+Authority Broadleaf- EPA Registration No. 279-3179 
+Sp~rtan - EPA Registration No. 279-3189 
+Authority First - EPA Registration No. 279-3195 
+Sp-artan 4F - EPA Registration No. 279-3220 
+Gauntlet CP - EPA Registration No. 279-3231 
+Suifentrazone Row- EPA Registration No. 279-3282 
+Dismiss - EPA Registration No. 279-3295 
+Echelon 0.3% Sand 
+EPA Registration No. 279-3317 
+Echelon 0.9% Granular Herbicide 
+EPA Registration No. 279-3318 
+Echelon 0.9% Granular Herbicide on Fertilizer 
+EPA Registration No. 279-3319 
+Echelon 0.57% Granular Herbicide on Fertilizer 
+EPA Registration No. 279-3320 
+Echelon 0.3% Granular Herbicide on Fertilizer 
+EPA Registration No. 279-3322 
+Echelon 4SC Herbicide 
+EPA Registration No. 279-3323 
+Echelon MUP Herbicide 
+EPA Registration No. 279-3324 
+Echelon 0.16% Granular Herbicide on Fertilizer 
+EPA Registration No. 279-3325 
+FEB 1 i 2010
+
+We have reviewed the regulatory files of the above named products and located 
+submitted data and documents that support rescinding the expiration dates. The 
+products are now active procjuct registrations. With this letter, we acknowledge 
+that the prior letter and email advising FMC of the expiration of these products, 
+dated July 21,2009 and August 31,2009, respectively, are withdrawn. 
+Enclosure 
+Sin/~rely yours,. . ~~. . 
+:;!)~) V ,m6\W;~ 
+Kathryn V. Montague 
+Product Manager (23) 
+Herbicide Branch 
+Registration Division (7505P)