feat: Introduce product scoring based on weighted sales metrics and store participation, integrated into the product grid API, and add a performance test script.

scripts/repro-500.js

@@ -0,0 +1,39 @@
+const { computeProductScores } = require('./src/lib/score-engine');
+
+// Simulate 50,000 rows
+const largeDataset = Array.from({ length: 50000 }, (_, i) => ({
+    codein: `item-${i}`,
+    codeFournisseur: 'A032',
+    nomFournisseur: 'ACCESS DECO',
+    libelle1: `Product ${i}`,
+    gtin: '123456789',
+    reference: 'REF',
+    code1: '20',
+    libelleNiveau1: 'LIQUIDES',
+    code2: '2001',
+    libelleNiveau2: 'EAUX',
+    code3: '200101',
+    libelle3: 'EAU SOURCE',
+    codeGamme: 'A',
+    codeGammeInit: 'A',
+    codeGammeDraft: null,
+    sales12m: {},
+    totalQuantite: Math.random() * 1000,
+    totalCa: Math.random() * 5000,
+    totalMarge: Math.random() * 1000,
+    tauxMarge: 20,
+    score: 0,
+    workingStores: ['MAG1', 'MAG2']
+}));
+
+console.log(`Starting test with ${largeDataset.length} rows...`);
+
+try {
+    const start = Date.now();
+    const result = computeProductScores(largeDataset);
+    const end = Date.now();
+    console.log(`Success! Processed ${result.length} rows in ${end - start}ms.`);
+} catch (error) {
+    console.error('Failed!', error);
+    process.exit(1);
+}