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/264-778.md

@@ -0,0 +1,97 @@
+# STRATEGO
+
+- EPA Reg No: **264-778**
+- Registrant: BAYER CROPSCIENCE, LLC
+- Signal word: Caution
+- Active ingredients: Propiconazole (20.9%); Trifloxystrobin (25%)
+- Label accepted: 2013-02-28
+- Source PDF: https://www3.epa.gov/pesticides/chem_search/ppls/000100-00948-20130228.pdf
+
+---
+-UNITED TATES ENVIRONMENTAL PROTECTION^ JENCY I
+C
+C.
+Steven A. Brooker
+Brooker Chemical Corporation
+9175 Eton Avenue
+Chatsworth, CA91311
+Dear Mr. Brooker:
+NOV 1 2 ZOOS
+Subject: Label Amendments •-
+EPAReg. No. 8281-1 (Hormex Rooting Powder No. 8)
+EPA Reg, No. 8281 -2 (Hormex Rooting Powder No. 16)
+EPA Reg. No, 8281-3 (Hormex Rooting Powder No. 3)
+EPA Reg. No. 8281-4 (Hormex Rooting Powder No. 30)
+EPA Reg. No, 8281 -5 (Hormex Rooting Powder No. 45)
+EPA Reg. No. 8281 -6 (Hormex Rooting Powder No. 1)
+Your .submissions dated September 28, 2009
+DecisionNos.: 370737, 370736, 370738, 370735, 370734, and 369315 ,
+The amendment referred to above submitted in connection with registration under
+the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), as amended, has been
+received and reviewed, and the amendment is acceptable provided that you submit and/or
+cite all data required for registration/reregistration of your product under FIFRA section
+3(c)(5) when the Agency requires all registrants of similar products to submit such data.
+If these conditions are not complied with, the registration will be subject to
+cancellation in accordance with FIFRA section 6(e). Your release for shipment of the
+product bearing the amended labeling constitutes acceptance of these conditions. Please
+submit three (3) copies of your final printed labeling prior to shipment of your product.
+Should you have any questions, you may contact Mr. Colin Walsh directly at (703) 308-
+0298 or via email at walsh.colin@epa.gov.
+Sincerely,
+i-/^
+Linda A. Hollis, Chief
+Biochemical Pesticides Branch
+Biopesticides and Pollution
+CONCURRENCES ^cvtiiuuii i^iviaiun V/J 1 11 ;
+;YMBOL ^
+iURNAME ^
+IATE ^
+^611 P
+J3S&.L.\\h,l<fi
+.isuf.
+c^f""I""/
+ilW*fPA Form 1320-1A (1/90) Printed on Recycled Paper OFFICIAL FILE COPY
+"' U.S. Government Printing Office: 2005 ^06-889 (mac)
+
+11111° a-p ™ «
+S3 -1s
+!«ifffltHifi
+jlllfjfllltfj
+illjfi|||llll
+.=l|!lli!l!s°^iiss» 8s11.?
+f £ 3 2 « s
+111 III
+Sif
+III-
+?l I» S"« ;
+: lli
+illli
+till
+i
+8 6
+If
+II
+aass
+;s- m a *•» i^ajsaiea
+il iliilfllliil
+il SS5?S5
+Is
+sS
+fl
+§1
+£3
+i
+ill
+3£li£|il
+ifljl
+°a ^ ™fB^ *)'!'*
+:i2*
+llli
+; as s
+hsi
+iii HI
+QUJ
+feLU
+uU
+o
+~