fix(rutracker): make the Cloudflare path actually work

[dethlex]

Closes #425

[Egorkat]

While testing this fix, found that parse_url() was missed — it still makes a bare request with no cf_clearance/User-Agent, so Cloudflare blocks it and the page has no <h1>, causing a silent "CantParse" failure. This breaks the "Add Torrent by URL" preview/validation flow (and likely periodic topic re-checks) even with this PR applied.

Suggested addition on top of this PR:

     def parse_url(self, url):
         match = self._regex.match(url)
         if match is None:
             return None

-        r = requests.get(url, allow_redirects=False, **self.tracker_settings.get_requests_kwargs())
+        r = requests.get(url, allow_redirects=False, headers=self.headers, cookies=self.cookies,
+                          **self.tracker_settings.get_requests_kwargs())

         soup = get_soup(r.text)

[Egorkat]

And in RutrackerPlugin:

     def parse_url(self, url):
+        # cf_clearance is required here too, or Cloudflare returns a challenge page with no <h1>
+        with DBSession() as db:
+            cred = db.query(self.credentials_class).first()
+            headers = parse_headers_field(cred.headers) if cred else None
+            cookies = parse_cookies_field(cred.cookies) if cred else None
+        self.tracker.headers = headers or {}
+        self.tracker.cookies = cookies or {}
         return self.tracker.parse_url(url)