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/10182-112.md

@@ -0,0 +1,62 @@
+# ORTHO GRAMOXONE CL
+
+- EPA Reg No: **10182-112**
+- Registrant: SYNGENTA CROP PROTECTION
+- Signal word: Danger
+- Active ingredients: Paraquat dichloride (29.1%)
+- Label accepted: 1986-06-11
+- Source PDF: https://www3.epa.gov/pesticides/chem_search/ppls/000239-02286-19860611.pdf
+
+---
+• 
+, 
+l 
+---- -'- - -- -~-~----------------------------..... 
+--
+Chevron 
+=== 
+PO~So)~ 
+DO NOT POUR 
+FROM ORIGINAL CONTAINER 
+EXCEPT FOR IMMEDIATE USE. 
+ONE SWALLOW CAN Kill! 
+- - --- .- - -- ~ - - ~ 
+, , 
+'GRAMOXONE' CL 
+--_.- - ._--- ._._._---- ----
+SUPPLEMENTAL PAMPHLET 
+DIRECTIONS FOR USE. READ WORKER SAFETY RULES ON BACK PANEL 
+TO' PREVENT ACCIDENTSr NEVER FYUT ~NTO FODOr 
+Dl,JN[( OR OTHER CONT~-.f!\fER~-:, 
+DO NOT USE OR STORE IN OR AROUND THE HOME 
+KEEP OUT OF REACH OF CHILDREN 
+A Weed and Grass Killer· Harvest Aid Chemical. 
+Active Ingredient 8)' VI: 
+:"',: ..... : ":'lJ'lde {~.1··dlfl1eth\'I·.J-: .1.: " . ?'" 
+Inert Ingredients 
+(\Jnt,1Ifl, ::> Ihe.. P:lLHllut r"ltl('l'l "." .'. ". ; .• '.J, ') 11'~" :''10 - ~ ......... - .... , .. ~ •. ~.'1~.~ ...... ~,.,~ I· ", •. :~~1 c. ~I c.. J(I,} 
+If SWALI OWE IJ 
+rrr· fR. [\ r~ [, lU \ LJ !I!~\\' \~J~ r; 't 
+CAN KILL IF SWALLOWED 
+HARMFUL TO THE EYES AND SKIN 
+--- -- ' .. -- - .. _._.,---_.- - -- _. __ . 
+IIRS' Alil TR[ATMENT 
+r Y[ !.[JNTAMINr, TION 
+. I,'.' .. ," .,'. " . . . 
+,I ". ',' • INItALA II(HI 
+'j 
+SKIN LOtI! AMINfl T ION 
+, , '. '. , :. " : .. ', '. 
+:. . ,', . , . . .'. . . 
+'" • ,;, I '. 
+-- ---- --- --, -;--Z't -.; . t 
+: l 
+I 
+I 
+II . , . I 
+I' 
+I: 
+I 
+) : 
+I 
+WASH AND DESTROY CONTAINER WHEN EMPTY. NEVER REUSE.