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/241-385.md

@@ -0,0 +1,70 @@
+# PURSUIT W HERBICIDE
+
+- EPA Reg No: **241-385**
+- Registrant: BASF AGRICULTURAL SOLUTIONS US LLC
+- Signal word: Caution
+- Active ingredients: Imazethapyr, ammonium salt (22.87%)
+- Label accepted: 2001-04-03
+- Source PDF: https://www3.epa.gov/pesticides/chem_search/ppls/000241-00385-20010403.pdf
+
+---
+EPA Reg. No. 241-385 
+SPLIT OUT USE DIRECTIONS TO ADD AN APPLICATION METHOD AND MODIFY 
+TIMINGS FOR USE IN SUCCULENT PEAS, DRY EDIBLE PEAS, 
+LENTILS, CHICKPEAS, AND LIMA BEANS 
+IN THE STATES OF WASHINGTON, IDAHO, OREGON AND MONT A:\ A ONLY 
+THIS LABEL EXPIRES APRIL 30, 2002 
+OBSERVE ALL DISCLAIMER AND PRECAUTIONARY STATEMENTS AND SPR-\YING, 
+APPLICATION, AND ROTATIO:-;AL CROP INFORMATION IN THE PURSUIT W LEAFLET LABEL 
+BEFORE USING. APPLY WITH GROUND EQUIPMENT ONLY. 
+DIRECTIONS FOR USE 
+It is a violation of Federal law to use this product in a 
+manner inconsistent with its labeling. 00 NOT 
+apply this product through any type of irrigation 
+system. This labeling must be ill the possession of 
+the user at the time of herbicide application. 
+USE RATE 
+Apply PURSUIT W herbicide at a broadcast rate of 3 
+ounces per acre for preplant, preplant incorporated. 
+and preemergence applications. 
+APPLICATION INSTRUCTIONS 
+Preplant and Preplant Incorporated Applications: 
+Apply PURSUIT W at the broadcast rate of 3 
+ounces per acre within 30 days before planting. If 
+incorporated, 00 NOT incorporate deeper than 3 
+inches. 
+In no-till and minimum tillage systems, PURSmT 
+W may be applied in the fall prior to spring 
+planting. Rainfall is required for incorporation. 
+Unpredictable weed control can be expected since 
+factors such as length of time between application 
+and planting as well as uncontrollable weather 
+factors (temperature, rainfall) will determine 
+herbicide activity and longevity. Apply PURSmT 
+W in the rail when soil temperatures at the 4-inch 
+depth are less than 55°F and before the ground is 
+frozen. 
+Preemergence Applications: 
+Apply PURSUIT W at the broadcast rate of 3 
+ounces per acre after planting, but prior to crop 
+emergence. 
+PURSUIT W may be tank-mixed with Sencor l 
+OF or Lexone 2 OF to assist in the control of 
+lambsquarters or mayweed-chamomile (dog­
+fennel). Refer to the Sencor or Lexone label for 
+proper application rates and restrictions. Always 
+use in accordance with the more restrictive label 
+restrictions and precautions. 
+CONSULT THE PURSUIT W LEAFLET LABEL 
+FOR OTHER PRECAUTIONS. 
+BASF Corporation 
+26 Davis Drive 
+Research Triangle Park, NC 27709 
+BASF 
+®Registcrcd Trademark ofBASF. 
+I Sencor is a trademark. of Bayer Corporation. 
+2Lexone is a trademark ofE.1. duPont de Nemours & Co. Inc. 
+©1001 BASF Corporation 
+All nghts reserved 
+NV:\. 2001-04-130-036 
+PE-17424 Rev. I 04/0301