feat(agent): scrape linked pages to ground answers in real content

Adds `_fetch_page_content()`, which downloads a URL with a
browser-like User-Agent and uses BeautifulSoup with the lxml
parser to strip scripts, navigation, and other non-content
markup, returning up to 3000 characters of readable text.

The response validator now also compares generated replies
against this scraped text, flagging named entities and claims
that do not appear in the retrieved page content.

Dependencies: adds beautifulsoup4==4.12.3 and pins lxml==5.3.0
for parser compatibility.

venv/lib/python3.11/site-packages/lxml/schematron.pxi

@@ -69,6 +69,9 @@ cdef class Schematron(_Validator):
     """
     cdef schematron.xmlSchematron* _c_schema
     cdef xmlDoc* _c_schema_doc
+    def __cinit__(self):
+        self._c_schema = NULL
+        self._c_schema_doc = NULL
 
     def __init__(self, etree=None, *, file=None):
         cdef _Document doc
@@ -80,14 +83,6 @@ cdef class Schematron(_Validator):
         if not config.ENABLE_SCHEMATRON:
             raise SchematronError, \
                 "lxml.etree was compiled without Schematron support."
-
-        import warnings
-        warnings.warn(
-            "The (non-ISO) Schematron feature is deprecated and will be removed from libxml2 and lxml. "
-            "Use 'lxml.isoschematron' instead.",
-            DeprecationWarning,
-        )
-
         if etree is not None:
             doc = _documentOrRaise(etree)
             root_node = _rootNodeOrRaise(etree)