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/5481-598.md

@@ -0,0 +1,55 @@
+# SCEPTER HERBICIDE CONTAINS SURFACTANT
+
+- EPA Reg No: **5481-598**
+- Registrant: AMVAC CHEMICAL CORPORATION
+- Signal word: Caution
+- Active ingredients: 3-Quinolinecarboxylic acid, 2-(4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazol-2-yl)-, monoammonium salt (17.3%)
+- Label accepted: 1987-08-26
+- Source PDF: https://www3.epa.gov/pesticides/chem_search/ppls/000241-00292-19870826.pdf
+
+---
+EPA Reg. No. 241-292 
+FOR USE IN aOYBEANS 
+OBSERVE ALL PRECAUTIONARY STATEMENTS IN THE 
+SCEPTER LEAFLET LABEL BEFORE USING 
+ROTATIONAL CROP RESTRICTIONS 
+The following rotational crops may be planted after applying SCEPTER at 
+recommended rates in soybeans: 
+1. Four months after last SCEPlER application: 
+Small Grains 
+Rice 
+2. Eleven months after last SCEPTER application: 
+Corn 
+Edible beans 
+Grain sorghum 
+Peanuts 
+Tobacco 
+3. Eighteen months after last SCEPTER application: 
+\ Cotton 
+ACCEPTED 
+AUG 261987 
+Undor the FedelollnsecliclrJo. 
+Fungicide. and R"denticj.j;.: ;:cl, i 
+aa umended. for the pe~licid8 
+, •• ;,t.,.d und,ojr// ~ "JCf I 
+EPA Reg. Nod)'-i 0'-/(7'-
+Only rotational crops harvested at maturity may be used for feed or food. 
+DO NOT plant other rotational crops prior to 18 months after last SCEPTER 
+application. 
+DO NOT apply Classic, Canopy, or Geminil the same year 
+injury to follow crops may O<;C,UI'. 
+I 
+In the event of a crop 10SB due to weather conditions, 
+replanted. DO NOT wr .:( the Boil deeper than 2 inches. 
+as SCEPTER or 
+•••••• Boybeaps. -;an lie' : .. . • ••••• • • .. 
+DO NOT GRAZE OR FEErJ TREATED SOYBEAN FORAGE, HAY OR STRAW TO :::.n1l::STOCOx.... . ( . . ••••• --------------------~----------------------------------------~I~~~--~~.--.. lClassic, Canopy, and Gemini are trademarks of E. I. du Pont de.Nemo~~~': 
+and Company. •• • •• 
+fIIIIII CYANAMID 
+\1----- . CiM"-~_ 
+t (( •••••• 
+(I ( •• 
+II II 
+, ( ( ( 
+, . 
+"