✨(backend) use pydantic AI to manage vercel data stream protocol

The frontend application is using Vercel AI SDK and it's data stream
protocol. We decided to use the pydantic AI library to use it's vercel
ai adapter. It will make the payload validation, use AsyncIterator and
deal with vercel specification.

src/backend/core/api/serializers.py

@@ -833,39 +833,6 @@ class AITranslateSerializer(serializers.Serializer):
         return value
 
 
-class AIProxySerializer(serializers.Serializer):
-    """Serializer for AI proxy requests."""
-
-    messages = serializers.ListField(
-        required=True,
-        child=serializers.DictField(),
-        allow_empty=False,
-    )
-    model = serializers.CharField(required=True)
-
-    def validate_messages(self, messages):
-        """Validate messages structure."""
-        # Ensure each message has the required fields
-        for message in messages:
-            if (
-                not isinstance(message, dict)
-                or "role" not in message
-                or "content" not in message
-            ):
-                raise serializers.ValidationError(
-                    "Each message must have 'role' and 'content' fields"
-                )
-
-        return messages
-
-    def validate_model(self, value):
-        """Validate model value is the same than settings.AI_MODEL"""
-        if value != settings.AI_MODEL:
-            raise serializers.ValidationError(f"{value} is not a valid model")
-
-        return value
-
-
 class MoveDocumentSerializer(serializers.Serializer):
     """
     Serializer for validating input data to move a document within the tree structure.