Fixed WA's Inf. Loop msgs. Validated Claude's msg_hist

Main changes:

fix: Modify main_whatsapp to handle incoming messages with background tasks for better performance

feat: Implement message splitting functionality in WhatsAppPresenter to handle long messages

fix: Added validation check to AnsariClaude to ensure that a "tool_use" block  is followed by only 1 "tool_result" block (in self.message_history)

Extra details:

fix: Added \n\n in AnsariClaude's response to separate between the text before tool's output and the tool's output

refactor: Remove unused `stream` option

feat: Enhance logging capabilities in ansari_logger to support file logging in DEV_MODE

docs: Add example JSON structures for API responses in documentation

feat: Update .gitignore to include logs directory

fix: Update Python version in Jupyter notebook for API response structure

Author: OdyAsh

.gitignore

@@ -10,6 +10,7 @@ docs/recordings/*
 etc/
 example_projects/
 lib/
+logs/
 share/
 src/ansari_backend.egg-info/*
 tmp/

docs/structure_of_api_responses/anthropic_api_structure_of_message_history.json

@@ -0,0 +1,56 @@
+[ // This is a dummy example, so don't focus on the content of the messages
+  {
+    "role": "user",
+    "content": "Can you search for information about machine learning?"
+  },
+  {
+    "role": "assistant",
+    "content": [
+      {
+        "type": "text",
+        "text": "I'll search for information about machine learning for you."
+      },
+      {
+        "type": "tool_use",
+        "id": "tool-123456",
+        "name": "search_database",
+        "input": {
+          "query": "machine learning",
+          "limit": 3
+        }
+      }
+    ]
+  },
+  {
+    "role": "user",
+    "content": [
+      {
+        "type": "tool_result",
+        "tool_use_id": "tool-123456",
+        "content": "{\"results\": [{\"title\": \"Introduction to ML\", \"url\": \"https://example.com/ml1\"}, {\"title\": \"Deep Learning Fundamentals\", \"url\": \"https://example.com/ml2\"}]}"
+      }
+    ]
+  },
+  {
+    "role": "assistant",
+    "content": [
+      {
+        "type": "text",
+        "text": "Based on my search, here are some resources about machine learning:\n\n1. Introduction to ML - This resource covers the basics of machine learning algorithms and techniques.\n2. Deep Learning Fundamentals - This focuses on neural networks and deep learning approaches."
+      }
+    ]
+  },
+  {
+    "role": "user",
+    "content": "Can you explain neural networks in more detail?"
+  },
+  {
+    "role": "assistant",
+    "content": [
+      {
+        "type": "text",
+        "text": "Neural networks are computational models inspired by the human brain. They consist of layers of interconnected nodes or \"neurons\" that process information.\n\nThe basic structure includes:\n\n1. Input layer: Receives initial data\n2. Hidden layers: Perform computations on inputs\n3. Output layer: Produces the final result\n\nEach connection between neurons has a weight that gets adjusted during training[1]."
+      }
+    ]
+  }
+]

docs/structure_of_api_responses/anthropic_api_structure_of_response.json

@@ -0,0 +1,20 @@
+{
+  "type": "content_block_start | content_block_delta | content_block_stop | message_delta | message_stop",
+  "content_block": {
+    "type": "text | tool_use",
+    "id": "tool-123456",
+    "name": "search_database"
+  },
+  "delta": {
+    "text": "Here's information about your query...",
+    "partial_json": "{ \"query\": \"ma",
+    "type": "citations_delta",
+    "citation": {
+      "start": 23,
+      "end": 45,
+      "number": 1,
+      "text": "according to source X..."
+    },
+    "stop_reason": "end_turn | tool_use"
+  }
+}

docs/structure_of_api_responses/openai_api_structure_of_chat_completion_chunk_object.ipynb

@@ -110,7 +110,7 @@
   },
   "language_info": {
    "name": "python",
-   "version": "3.12.5"
+   "version": "3.13.2"
   }
  },
  "nbformat": 4,

src/ansari/agents/ansari.py

@@ -95,7 +95,7 @@ def process_input(self, user_input: str):
                 if m:
                     yield m
 
-    def replace_message_history(self, message_history: list[dict], use_tool=True, stream=True):
+    def replace_message_history(self, message_history: list[dict], use_tool=True):
         """
         Replaces the current message history (stored in Ansari) with the given message history,
         and then processes it to generate a response from Ansari.
@@ -106,23 +106,19 @@ def replace_message_history(self, message_history: list[dict], use_tool=True, st
             {"role": "system", "content": self.sys_msg},
         ] + message_history
 
-        # Return/Yield Ansari's response to the user
-        # TODO(odyash) later (good_first_issue): `stream == False` is not implemented yet; so it has to stay `True`
-        for m in self.process_message_history(use_tool, stream=stream):
+        # Yield Ansari's response to the user
+        for m in self.process_message_history(use_tool):
             if m:
                 yield m
 
     def get_completion(self, **kwargs):
         return litellm.completion(**kwargs)
 
-    def process_message_history(self, use_tool=True, stream=True):
-        """
-        TODO(odyash) later (good_first_issue): `stream == False` is not implemented yet; so it has to stay `True`
-        """
+    def process_message_history(self, use_tool=True):
         common_params = {
             "model": self.model,
             "messages": self.message_history,
-            "stream": stream,
+            "stream": True,
             "stream_options": {"include_usage": True},
             "timeout": 30.0,
             "temperature": 0.0,
@@ -225,14 +221,11 @@ def process_message_history(self, use_tool=True, stream=True):
         else:
             raise Exception("Invalid response mode: " + response_mode)
 
-    def process_one_round(self, use_tool=True, stream=True):
-        """
-        TODO(odyash) later (good_first_issue): `stream == False` is not implemented yet; so it has to stay `True`
-        """
+    def process_one_round(self, use_tool=True):
         common_params = {
             "model": self.model,
             "messages": self.message_history,
-            "stream": stream,
+            "stream": True,
             "stream_options": {"include_usage": True},
             "timeout": 30.0,
             "temperature": 0.0,