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/includes/tree.pxd

@@ -6,9 +6,8 @@ cdef extern from "lxml-version.h":
     cdef char* LXML_VERSION_STRING
 
 cdef extern from "libxml/xmlversion.h":
-    const char* xmlParserVersion
-    int LIBXML_VERSION
-
+    cdef const_char* xmlParserVersion
+    cdef int LIBXML_VERSION
 
 cdef extern from "libxml/xmlstring.h" nogil:
     ctypedef unsigned char xmlChar
@@ -142,7 +141,7 @@ cdef extern from "libxml/tree.h" nogil:
         XML_ATTRIBUTE_NMTOKENS=    8
         XML_ATTRIBUTE_ENUMERATION= 9
         XML_ATTRIBUTE_NOTATION=    10
-
+    
     ctypedef enum xmlAttributeDefault:
         XML_ATTRIBUTE_NONE=     1
         XML_ATTRIBUTE_REQUIRED= 2
@@ -289,7 +288,7 @@ cdef extern from "libxml/tree.h" nogil:
         xmlDtd* intSubset
         xmlDtd* extSubset
         int properties
-
+        
     ctypedef struct xmlAttr:
         void* _private
         xmlElementType type
@@ -308,7 +307,7 @@ cdef extern from "libxml/tree.h" nogil:
         const_xmlChar* name
         xmlAttr* attr
         xmlDoc* doc
-
+        
     ctypedef struct xmlBuffer
 
     ctypedef struct xmlBuf   # new in libxml2 2.9
@@ -319,14 +318,14 @@ cdef extern from "libxml/tree.h" nogil:
         int error
 
     const_xmlChar* XML_XML_NAMESPACE
-
+        
     cdef void xmlFreeDoc(xmlDoc* cur)
     cdef void xmlFreeDtd(xmlDtd* cur)
     cdef void xmlFreeNode(xmlNode* cur)
     cdef void xmlFreeNsList(xmlNs* ns)
     cdef void xmlFreeNs(xmlNs* ns)
     cdef void xmlFree(void* buf)
-
+    
     cdef xmlNode* xmlNewNode(xmlNs* ns, const_xmlChar* name)
     cdef xmlNode* xmlNewDocText(xmlDoc* doc, const_xmlChar* content)
     cdef xmlNode* xmlNewDocComment(xmlDoc* doc, const_xmlChar* content)
@@ -438,7 +437,7 @@ cdef extern from "libxml/xmlIO.h":
     cdef xmlOutputBuffer* xmlOutputBufferCreateIO(
         xmlOutputWriteCallback iowrite,
         xmlOutputCloseCallback ioclose,
-        void * ioctx,
+        void * ioctx, 
         xmlCharEncodingHandler* encoder) nogil
     cdef xmlOutputBuffer* xmlOutputBufferCreateFile(
         stdio.FILE* file, xmlCharEncodingHandler* encoder) nogil
@@ -472,11 +471,14 @@ cdef extern from "libxml/globals.h" nogil:
     cdef int xmlThrDefKeepBlanksDefaultValue(int onoff)
     cdef int xmlThrDefLineNumbersDefaultValue(int onoff)
     cdef int xmlThrDefIndentTreeOutput(int onoff)
-
+    
 cdef extern from "libxml/xmlmemory.h" nogil:
     cdef void* xmlMalloc(size_t size)
     cdef int xmlMemBlocks()
     cdef int xmlMemUsed()
+    cdef void xmlMemDisplay(stdio.FILE* file)
+    cdef void xmlMemDisplayLast(stdio.FILE* file, long num_bytes)
+    cdef void xmlMemShow(stdio.FILE* file, int count)
 
 cdef extern from "etree_defs.h" nogil:
     cdef bint _isElement(xmlNode* node)