Add Record/Replay functionality for offline processing (Issue #2759)

Co-Authored-By: Joe Moura <joao@crewai.com>

tests/llm_response_cache_test.py

@@ -0,0 +1,78 @@
+import pytest
+from unittest.mock import MagicMock, patch
+
+from crewai.llm import LLM
+from crewai.utilities.llm_response_cache_handler import LLMResponseCacheHandler
+
+
+@pytest.fixture
+def handler():
+    handler = LLMResponseCacheHandler()
+    handler.storage.add = MagicMock()
+    handler.storage.get = MagicMock()
+    return handler
+
+
+@pytest.mark.vcr(filter_headers=["authorization"])
+def test_llm_recording(handler):
+    handler.start_recording()
+    
+    llm = LLM(model="gpt-4o-mini")
+    llm.set_response_cache_handler(handler)
+    
+    messages = [{"role": "user", "content": "Hello, world!"}]
+    
+    with patch('litellm.completion') as mock_completion:
+        mock_completion.return_value = {
+            "choices": [{"message": {"content": "Hello, human!"}}]
+        }
+        
+        response = llm.call(messages)
+        
+        assert response == "Hello, human!"
+        
+        handler.storage.add.assert_called_once_with(
+            "gpt-4o-mini", messages, "Hello, human!"
+        )
+
+
+@pytest.mark.vcr(filter_headers=["authorization"])
+def test_llm_replaying(handler):
+    handler.start_replaying()
+    handler.storage.get.return_value = "Cached response"
+    
+    llm = LLM(model="gpt-4o-mini")
+    llm.set_response_cache_handler(handler)
+    
+    messages = [{"role": "user", "content": "Hello, world!"}]
+    
+    with patch('litellm.completion') as mock_completion:
+        response = llm.call(messages)
+        
+        assert response == "Cached response"
+        
+        mock_completion.assert_not_called()
+        
+        handler.storage.get.assert_called_once_with("gpt-4o-mini", messages)
+
+
+@pytest.mark.vcr(filter_headers=["authorization"])
+def test_llm_replay_fallback(handler):
+    handler.start_replaying()
+    handler.storage.get.return_value = None
+    
+    llm = LLM(model="gpt-4o-mini")
+    llm.set_response_cache_handler(handler)
+    
+    messages = [{"role": "user", "content": "Hello, world!"}]
+    
+    with patch('litellm.completion') as mock_completion:
+        mock_completion.return_value = {
+            "choices": [{"message": {"content": "Hello, human!"}}]
+        }
+        
+        response = llm.call(messages)
+        
+        assert response == "Hello, human!"
+        
+        mock_completion.assert_called_once()