discover: fix recipe-scrapers 15.6 API — drop wild_mode kw

scrape_me(url) no longer accepts wild_mode in 15.x. Two-stage fallback:
  1. scrape_me(url) — site-specific scraper (best quality)
  2. fetch html + scrape_html(html, org_url, supported_only=False) —
     generic JSON-LD/microdata pass for unsupported sites

Caught at first dogfood test against allrecipes; previous code raised
"unexpected keyword argument 'wild_mode'" before issuing any HTTP.

cauldron/discover_recipes.py

@@ -146,10 +146,28 @@ def _scrape_one(url: str) -> tuple[dict, str | None] | None:
         return None
 
     try:
-        scraper = scrape_me(url, wild_mode=True)
+        scraper = scrape_me(url)
     except Exception as e:
-        log.warning("[discover] scrape_me(%s) failed: %s", url, e)
-        return None
+        # scraper_exists_for is False or the request failed.
+        # Fall back to scrape_html with supported_only=False so unknown
+        # sites still get a JSON-LD/microdata pass.
+        try:
+            from recipe_scrapers import scrape_html  # type: ignore
+            import requests as _rq
+            resp = _rq.get(
+                url,
+                timeout=15,
+                headers={"User-Agent": "Mozilla/5.0 (cauldron-discover)"},
+            )
+            if resp.status_code != 200:
+                log.warning("[discover] fetch %s -> %s", url, resp.status_code)
+                return None
+            scraper = scrape_html(
+                html=resp.text, org_url=url, supported_only=False
+            )
+        except Exception as e2:
+            log.warning("[discover] scrape_me(%s) failed: %s / fallback: %s", url, e, e2)
+            return None
 
     shaped = _to_mealie_shape(scraper, url)
     image = shaped.get("image") or None