feat(parser): replacing tldextract with tld library

This might break the adaptive data users have for websites BUT:
1. tld uses ~3.7x less memory during extraction operations (1.5 MB vs 5.7 MB).
2. tld uses ~56% less memory on import (5.2 MB vs 11.9 MB).
3. Zero dependencies (vs 3 for tldextract).

In return, it's 30ms slower for extracting 5000 URLs, which is negligible. Also, the type hints aren't always accurate, but it's fine; I corrected them.

scrapling/engines/toolbelt/fingerprints.py

@@ -5,11 +5,11 @@ Functions related to generating headers and fingerprints generally
 from functools import lru_cache
 from platform import system as platform_system
 
-from tldextract import extract
+from tld import get_tld, Result
 from browserforge.headers import Browser, HeaderGenerator
 from browserforge.headers.generator import SUPPORTED_OPERATING_SYSTEMS
 
-from scrapling.core._types import Dict, Literal, Tuple
+from scrapling.core._types import Dict, Literal, Tuple, cast
 
 __OS_NAME__ = platform_system()
 OSName = Literal["linux", "macos", "windows"]
@@ -28,11 +28,15 @@ def generate_convincing_referer(url: str) -> str | None:
     :param url: The URL you are about to fetch.
     :return: Google's search URL of the domain name, or None for localhost/IP addresses
     """
-    extracted = extract(url)
+    # Fixing the inaccurate return type hint in `get_tld`
+    extracted: Result | None = cast(Result, get_tld(url, as_object=True, fail_silently=True))
+    if not extracted:
+        return None
+
     website_name = extracted.domain
 
     # Skip generating referer for localhost, IP addresses, or when there's no valid domain
-    if not website_name or not extracted.suffix or website_name in ("localhost", "127.0.0.1", "::1"):
+    if not website_name or not extracted.tld or website_name in ("localhost", "127.0.0.1", "::1"):
         return None
 
     # Check if it's an IP address (simple check for IPv4)