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/html/__init__.py

@@ -70,7 +70,7 @@ _forms_xpath = etree.XPath("descendant-or-self::form|descendant-or-self::x:form"
 #_class_xpath = etree.XPath(r"descendant-or-self::*[regexp:match(@class, concat('\b', $class_name, '\b'))]", {'regexp': 'http://exslt.org/regular-expressions'})
 _class_xpath = etree.XPath("descendant-or-self::*[@class and contains(concat(' ', normalize-space(@class), ' '), concat(' ', $class_name, ' '))]")
 _id_xpath = etree.XPath("descendant-or-self::*[@id=$id]")
-_collect_string_content = etree.XPath("string()", smart_strings=False)
+_collect_string_content = etree.XPath("string()")
 _iter_css_urls = re.compile(r'url\(('+'["][^"]*["]|'+"['][^']*[']|"+r'[^)]*)\)', re.I).finditer
 _iter_css_imports = re.compile(r'@import "(.*?)"').finditer
 _label_xpath = etree.XPath("//label[@for=$id]|//x:label[@for=$id]",
@@ -263,9 +263,7 @@ class HtmlMixin:
         Return the <body> element.  Can be called from a child element
         to get the document's head.
         """
-        for element in self.getroottree().iter("body", f"{{{XHTML_NAMESPACE}}}body"):
-            return element
-        return None
+        return self.xpath('//body|//x:body', namespaces={'x':XHTML_NAMESPACE})[0]
 
     @property
     def head(self):
@@ -273,9 +271,7 @@ class HtmlMixin:
         Returns the <head> element.  Can be called from a child
         element to get the document's head.
         """
-        for element in self.getroottree().iter("head", f"{{{XHTML_NAMESPACE}}}head"):
-            return element
-        return None
+        return self.xpath('//head|//x:head', namespaces={'x':XHTML_NAMESPACE})[0]
 
     @property
     def label(self):