fix(og): broaden social-crawler coverage (Bluesky Cardyb, WeChat-scraper, fediverse, etc.)

From alexvaltchev's field UA list on #699. Adds CRAWLER-EXCLUSIVE tokens
to both the nginx UA regex and SOCIAL_CRAWLER_PATTERNS (kept in sync):
Cardyb (Bluesky's actual link-card fetcher), facebookcatalog, Signal,
Misskey, Pleroma, Synapse, Nextcloud, Rocket.Chat, kakaotalk-scrap,
Google-PageRenderer, OdklBot, ZoomBot.

Deliberately NOT added: UAs shared with real human in-app browsers
(WeChat MicroMessenger, LINE 'Line/', Zalo) and broad strings
('InAppBrowser', 'preview', 'unfurl', 'XING' → matches 'boxing'). Our OG
response is meta-only with no redirect, so matching those would serve a
human the bare stub. New negative test locks that exclusion in.

Verified: nginx -t passes; live harness confirms the new tokens rewrite
to /og while the in-app-browser UAs still get the SPA. Backend suite 15/15.

backend/src/__tests__/galleryOgService.shareImage.test.js

@@ -343,14 +343,34 @@ describe('isSocialCrawler — extended bot coverage (#521)', () => {
       // 3rd-party preview services used by business-messaging stacks
       'LinkPreview/1.0',
       'Slack-ImgProxy/1.0',
-      // Viber link-preview fetcher (#699 follow-up)
+      // Viber + broader crawler set (#699 follow-up)
       'Mozilla/5.0 (compatible; Viber)',
+      'Mozilla/5.0 (compatible; Bluesky Cardyb/1.1)',
+      'facebookcatalog/1.0',
+      'kakaotalk-scrap/1.0',
+      'Mozilla/5.0 (compatible; Synapse/1.98)',
+      'Rocket.Chat/6.0',
     ];
     for (const ua of knownBots) {
       expect(isSocialCrawler(ua)).toBe(true);
     }
   });
 
+  it('does NOT match human in-app-browser UAs (our OG response is meta-only, no redirect)', () => {
+    // These share a token with a preview bot but are also sent by real users
+    // browsing inside the app's webview — matching them would serve a human
+    // the bare OG stub. Deliberately excluded; guard against re-adding them.
+    const inAppBrowsers = [
+      'Mozilla/5.0 (iPhone) AppleWebKit MicroMessenger/8.0.0',        // WeChat in-app
+      'Mozilla/5.0 (iPhone) AppleWebKit Line/13.0.0',                // LINE in-app
+      'Mozilla/5.0 (Linux; Android) Zalo',                          // Zalo in-app
+      'Mozilla/5.0 (Macintosh) Chrome/120.0 Safari/537.36 boxing',  // "XING" substring trap
+    ];
+    for (const ua of inAppBrowsers) {
+      expect(isSocialCrawler(ua)).toBe(false);
+    }
+  });
+
   it('does not match a regular browser UA', () => {
     const browsers = [
       'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0 Safari/537.36',

backend/src/services/galleryOgService.js

@@ -38,7 +38,25 @@ const SOCIAL_CRAWLER_PATTERNS = [
   // Viber's link-preview fetcher — was never detected, so shared links
   // showed no rich preview in Viber (#699 follow-up). Keep in sync with the
   // UA list in frontend/nginx.conf.
-  /Viber/i
+  /Viber/i,
+  // Broader crawler coverage (#699 follow-up, from alexvaltchev's field list).
+  // IMPORTANT: only CRAWLER-EXCLUSIVE tokens are added here. Our OG response is
+  // meta-only (no client redirect), so a UA shared with a real human in-app
+  // browser would serve that human the bare stub. That rules out WeChat
+  // (MicroMessenger), LINE (Line/), Zalo, and generic strings like
+  // "InAppBrowser"/"preview"/"unfurl" — deliberately NOT added.
+  /Cardyb/i, // Bluesky's link-card service (the actual fetcher UA)
+  /facebookcatalog/i, // Facebook catalog crawler
+  /Signal/i, // Signal link preview
+  /Misskey/i, // fediverse (server-side preview fetch)
+  /Pleroma/i, // fediverse
+  /Synapse/i, // Matrix homeserver URL preview
+  /Nextcloud/i, // Nextcloud Talk/News link crawler
+  /Rocket\.Chat/i, // Rocket.Chat server preview
+  /kakaotalk-scrap/i, // KakaoTalk's scraper (NOT the in-app browser UA)
+  /Google-PageRenderer/i, // Google Chat previews (not Search)
+  /OdklBot/i, // Odnoklassniki
+  /ZoomBot/i // Zoom Team Chat
 ];
 
 function isSocialCrawler(userAgent) {