fix(setup,docs): rebuild locally built images, name OCR_DEEPSEEK_URL

Both setup scripts start with `compose pull && compose up -d`. In the local
compose file backend and worker are build-only services, so `up -d` builds
only when no image exists yet: a rerun that switches OCR on wrote
INSTALL_TESSERACT=true to .env and then reused the image built without it,
leaving OCR_ENABLED=true with no engine. Build explicitly on the local
compose path; the hub path stays pull-only, its services have no build stage.

The OCR message named OCR_ENGINE=deepseek but not OCR_DEEPSEEK_URL, whose
default (localhost:11434) resolves to the container, not the host.

deployment/sandbox/README.md still described Docling as already present in
application/requirements.txt.

deployment/sandbox/README.md

@@ -188,8 +188,10 @@ Daytona render output matches the Jupyter-backend output.
 
 Document reading no longer runs in this sandbox. The `read_document` tool and the
 workflow native-file extract branch enqueue a `parse_document` Celery task that
-parses the document **in the backend** (Docling, already in
-`application/requirements.txt`) and awaits the result. The task is routed to a
+parses the document **in the backend** (the `DOC_PARSER_ENGINE` parser — anydoc
+by default; Docling only when the optional
+`application/requirements-docling.txt` extra is installed) and awaits the
+result. The task is routed to a
 dedicated **`parsing` queue** (`settings.DOCUMENT_PARSE_QUEUE`, default
 `"parsing"`) so a parse enqueued from inside a Celery worker (headless/scheduled
 agent) is served by a separate worker and never self-deadlocks the awaiting one.

setup.ps1

@@ -523,7 +523,7 @@ function Configure-DocProcessing {
         # Bakes tesseract into the locally built images (docker compose
         # --env-file .env build).
         "INSTALL_TESSERACT=true" | Add-Content -Path $ENV_FILE -Encoding utf8
-        Write-ColorText "OCR enabled. tesseract will be built into the images (INSTALL_TESSERACT=true); set OCR_ENGINE=deepseek in .env for a DeepSeek-OCR endpoint instead." -ForegroundColor "Green"
+        Write-ColorText "OCR enabled. tesseract will be built into the images (INSTALL_TESSERACT=true); for a DeepSeek-OCR endpoint instead, set OCR_ENGINE=deepseek and OCR_DEEPSEEK_URL=<endpoint> in .env." -ForegroundColor "Green"
         $docling_ocr = Read-Host "Also install the Docling layout engine for OCR (better tables/reading order, several GB heavier)? (y/N)"
         if ($docling_ocr -eq "y" -or $docling_ocr -eq "Y") {
             # Locally built images include docling via this build arg; it becomes
@@ -647,7 +647,14 @@ function Use-DocsPublicAPIEndpoint {
             throw "Docker compose pull failed with exit code $LASTEXITCODE"
         }
         
-        & docker compose --env-file "$ENV_FILE" -f "$COMPOSE_FILE" up -d
+        # Locally built images must be rebuilt rather than reused:
+        # INSTALL_TESSERACT and INSTALL_DOCLING are build args, so a rerun that
+        # switches OCR on would otherwise keep the image built without them.
+        $up_args = @("up", "-d")
+        if ($COMPOSE_FILE -eq $COMPOSE_FILE_LOCAL) {
+            $up_args = @("up", "--build", "-d")
+        }
+        & docker compose --env-file "$ENV_FILE" -f "$COMPOSE_FILE" @up_args
         if ($LASTEXITCODE -ne 0) {
             throw "Docker compose up failed with exit code $LASTEXITCODE"
         }

setup.sh

@@ -377,7 +377,7 @@ configure_doc_processing() {
         # Bakes tesseract into the locally built images (docker compose
         # --env-file .env build).
         echo "INSTALL_TESSERACT=true" >> "$ENV_FILE"
-        echo -e "${GREEN}OCR enabled. tesseract will be built into the images (INSTALL_TESSERACT=true); set OCR_ENGINE=deepseek in .env for a DeepSeek-OCR endpoint instead.${NC}"
+        echo -e "${GREEN}OCR enabled. tesseract will be built into the images (INSTALL_TESSERACT=true); for a DeepSeek-OCR endpoint instead, set OCR_ENGINE=deepseek and OCR_DEEPSEEK_URL=<endpoint> in .env.${NC}"
         read -p "$(echo -e "${DEFAULT_FG}Also install the Docling layout engine for OCR (better tables/reading order, several GB heavier)? (y/N): ${NC}")" docling_ocr
         if [[ "$docling_ocr" =~ ^[yY]$ ]]; then
             # Locally built images include docling via this build arg; it becomes
@@ -469,7 +469,14 @@ use_docs_public_api_endpoint() {
     check_and_start_docker
 
     echo -e "\n${NC}Starting Docker Compose...${NC}"
-    docker compose --env-file "${ENV_FILE}" -f "${COMPOSE_FILE}" pull && docker compose --env-file "${ENV_FILE}" -f "${COMPOSE_FILE}" up -d
+    # Locally built images must be rebuilt rather than reused: INSTALL_TESSERACT
+    # and INSTALL_DOCLING are build args, so a rerun that switches OCR on would
+    # otherwise keep the image that was built without them.
+    up_args=(up -d)
+    if [ "$COMPOSE_FILE" = "$COMPOSE_FILE_LOCAL" ]; then
+        up_args=(up --build -d)
+    fi
+    docker compose --env-file "${ENV_FILE}" -f "${COMPOSE_FILE}" pull && docker compose --env-file "${ENV_FILE}" -f "${COMPOSE_FILE}" "${up_args[@]}"
     docker_compose_status=$? # Capture exit status of docker compose
 
     echo "Docker Compose Exit Status: $docker_compose_status"