Fix login-detection glob requiring a path segment after the community

wait_for_url("**/{community}/**") requires characters both before and
after the community slug. The community home page itself is exactly
https://www.skool.com/<community> with nothing trailing, so the first
real run navigated there correctly (confirmed in the logs: chase-ai
redirect, then /navaigate) but still timed out waiting for a pattern
that could never match. Switched to a plain substring predicate.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ViRGe7WHuUXUpHqvUfuW5r

scrape.py

@@ -91,7 +91,7 @@ async def run(community: str, output_dir: Path, discover: bool = False):
         print("Opening Skool — please log in when the browser window appears.")
         print("The script will continue automatically once you land on the community.\n")
         await page.goto("https://www.skool.com/login")
-        await page.wait_for_url(f"**/{community}/**", timeout=300_000)
+        await page.wait_for_url(lambda url: f"/{community}" in url, timeout=300_000)
         print("Logged in.\n")
 
         await page.goto(classroom)