feat: async product/receipt recognition via Kafka

Backend:
- Migration 002: product_recognition_jobs table with JSONB images column
  and job_type CHECK ('receipt' | 'products')
- New Kafka topics: ai.products.paid / ai.products.free
- ProductJob model, ProductJobRepository (mirrors dish job pattern)
- itemEnricher extracted from Handler — shared by HTTP handler and worker
- ProductSSEBroker: PG LISTEN on product_job_update channel
- ProductWorkerPool: 5 workers, branches on job_type to call
  RecognizeReceipt or RecognizeProducts per image in parallel
- Handler: RecognizeReceipt and RecognizeProducts now return 202 Accepted
  instead of blocking; 4 new endpoints: GET /ai/product-jobs,
  /product-jobs/history, /product-jobs/{id}, /product-jobs/{id}/stream
- cmd/worker: extended to run ProductWorkerPool alongside dish WorkerPool
- cmd/server: wires productJobRepository + productSSEBroker; both SSE
  brokers started in App.Start()

Flutter client:
- ProductJobCreated, ProductJobResult, ProductJobSummary, ProductJobEvent
  models + submitReceiptRecognition/submitProductsRecognition/stream methods
- Shared _openSseStream helper eliminates duplicate SSE parsing loop
- ScanScreen: replace blocking AI calls with async submit + navigate to
  ProductJobWatchScreen
- ProductJobWatchScreen: watches SSE stream, navigates to /scan/confirm
  when done, shows error on failure
- ProductsScreen: prepends _RecentScansSection (hidden when empty); compact
  horizontal list of recent scans with "See all" → history
- ProductJobHistoryScreen: full list of all product recognition jobs
- New routes: /scan/product-job-watch, /products/job-history
- L10n: 7 new keys in all 12 ARB files

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

backend/internal/domain/recognition/item_enricher.go

@@ -0,0 +1,98 @@
+package recognition
+
+import (
+	"context"
+	"log/slog"
+
+	"github.com/food-ai/backend/internal/adapters/ai"
+	"github.com/food-ai/backend/internal/domain/product"
+)
+
+// itemEnricher matches recognized items against the product catalog,
+// triggering AI classification for unknown items.
+// Extracted from Handler so both the HTTP handler and the product worker pool can use it.
+type itemEnricher struct {
+	recognizer  Recognizer
+	productRepo ProductRepository
+}
+
+func newItemEnricher(recognizer Recognizer, productRepo ProductRepository) *itemEnricher {
+	return &itemEnricher{recognizer: recognizer, productRepo: productRepo}
+}
+
+// enrich matches each recognized item against the product catalog.
+// Items without a match trigger a classification call and upsert into the DB.
+func (enricher *itemEnricher) enrich(enrichContext context.Context, items []ai.RecognizedItem) []EnrichedItem {
+	result := make([]EnrichedItem, 0, len(items))
+	for _, item := range items {
+		enriched := EnrichedItem{
+			Name:        item.Name,
+			Quantity:    item.Quantity,
+			Unit:        item.Unit,
+			Category:    item.Category,
+			Confidence:  item.Confidence,
+			StorageDays: 7, // sensible default
+		}
+
+		catalogProduct, matchError := enricher.productRepo.FuzzyMatch(enrichContext, item.Name)
+		if matchError != nil {
+			slog.WarnContext(enrichContext, "fuzzy match product", "name", item.Name, "err", matchError)
+		}
+
+		if catalogProduct != nil {
+			productID := catalogProduct.ID
+			enriched.MappingID = &productID
+			if catalogProduct.DefaultUnit != nil {
+				enriched.Unit = *catalogProduct.DefaultUnit
+			}
+			if catalogProduct.StorageDays != nil {
+				enriched.StorageDays = *catalogProduct.StorageDays
+			}
+			if catalogProduct.Category != nil {
+				enriched.Category = *catalogProduct.Category
+			}
+		} else {
+			classification, classifyError := enricher.recognizer.ClassifyIngredient(enrichContext, item.Name)
+			if classifyError != nil {
+				slog.WarnContext(enrichContext, "classify unknown product", "name", item.Name, "err", classifyError)
+			} else {
+				saved := enricher.saveClassification(enrichContext, classification)
+				if saved != nil {
+					savedID := saved.ID
+					enriched.MappingID = &savedID
+				}
+				enriched.Category = classification.Category
+				enriched.Unit = classification.DefaultUnit
+				enriched.StorageDays = classification.StorageDays
+			}
+		}
+		result = append(result, enriched)
+	}
+	return result
+}
+
+// saveClassification upserts an AI-produced classification into the product catalog.
+func (enricher *itemEnricher) saveClassification(enrichContext context.Context, classification *ai.IngredientClassification) *product.Product {
+	if classification == nil || classification.CanonicalName == "" {
+		return nil
+	}
+
+	catalogProduct := &product.Product{
+		CanonicalName:   classification.CanonicalName,
+		Category:        strPtr(classification.Category),
+		DefaultUnit:     strPtr(classification.DefaultUnit),
+		CaloriesPer100g: classification.CaloriesPer100g,
+		ProteinPer100g:  classification.ProteinPer100g,
+		FatPer100g:      classification.FatPer100g,
+		CarbsPer100g:    classification.CarbsPer100g,
+		StorageDays:     intPtr(classification.StorageDays),
+	}
+
+	saved, upsertError := enricher.productRepo.Upsert(enrichContext, catalogProduct)
+	if upsertError != nil {
+		slog.WarnContext(enrichContext, "upsert classified product", "name", classification.CanonicalName, "err", upsertError)
+		return nil
+	}
+
+	return saved
+}