feat: OG image scraper - auto-fetch recipe images from Pinterest/URLs

- New backend service: og-scraper.service.ts (extracts og:image, og:title, og:description)
- Pinterest support via Twitterbot UA (gets original resolution from i.pinimg.com)
- Works with Chefkoch, Allrecipes, blogs, any site with og:image meta tags
- GET /api/og-preview?url= for preview
- POST /api/recipes/:id/fetch-image to download + process with sharp
- Frontend: 'Bild holen' button appears when source URL is filled
- Auto-fills title & description from OG data if empty
- Images processed to WebP, max 1200px wide

backend/src/services/og-scraper.service.ts

@@ -0,0 +1,73 @@
+/**
+ * OG Image Scraper — holt og:image aus beliebiger URL
+ * Funktioniert mit Pinterest, Chefkoch, Allrecipes, Blogs etc.
+ */
+
+const USER_AGENTS: Record<string, string> = {
+  'pinterest.com': 'Twitterbot/1.0',
+  'pinterest.de': 'Twitterbot/1.0',
+  'pin.it': 'Twitterbot/1.0',
+  default: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
+};
+
+function getUserAgent(url: string): string {
+  for (const [domain, ua] of Object.entries(USER_AGENTS)) {
+    if (domain !== 'default' && url.includes(domain)) return ua;
+  }
+  return USER_AGENTS.default;
+}
+
+export interface OgData {
+  image?: string;
+  title?: string;
+  description?: string;
+}
+
+export async function scrapeOgData(url: string): Promise<OgData> {
+  const ua = getUserAgent(url);
+
+  const res = await fetch(url, {
+    headers: { 'User-Agent': ua },
+    redirect: 'follow',
+    signal: AbortSignal.timeout(10000),
+  });
+
+  if (!res.ok) throw new Error(`HTTP ${res.status}`);
+
+  const html = await res.text();
+  const result: OgData = {};
+
+  // Extract og:image
+  const imageMatch = html.match(/<meta[^>]+(?:property|name)="og:image"[^>]+content="([^"]+)"/i)
+    || html.match(/<meta[^>]+content="([^"]+)"[^>]+(?:property|name)="og:image"/i);
+  if (imageMatch) result.image = imageMatch[1];
+
+  // Extract og:title
+  const titleMatch = html.match(/<meta[^>]+(?:property|name)="og:title"[^>]+content="([^"]+)"/i)
+    || html.match(/<meta[^>]+content="([^"]+)"[^>]+(?:property|name)="og:title"/i);
+  if (titleMatch) result.title = decodeHtmlEntities(titleMatch[1]);
+
+  // Extract og:description
+  const descMatch = html.match(/<meta[^>]+(?:property|name)="og:description"[^>]+content="([^"]+)"/i)
+    || html.match(/<meta[^>]+content="([^"]+)"[^>]+(?:property|name)="og:description"/i);
+  if (descMatch) result.description = decodeHtmlEntities(descMatch[1]);
+
+  // Pinterest special: higher-res image from JSON data
+  if (url.includes('pinterest') && result.image) {
+    // Try to get orig resolution instead of 736x
+    const origMatch = html.match(/"orig":\s*\{[^}]*"url"\s*:\s*"([^"]+)"/);
+    if (origMatch) result.image = origMatch[1];
+  }
+
+  return result;
+}
+
+function decodeHtmlEntities(str: string): string {
+  return str
+    .replace(/&amp;/g, '&')
+    .replace(/&lt;/g, '<')
+    .replace(/&gt;/g, '>')
+    .replace(/&quot;/g, '"')
+    .replace(/&#39;/g, "'")
+    .replace(/&#x27;/g, "'");
+}