Add Ollama support

Related to Integuru-AI#7

Add support for Ollama model using the ollama python package.

* **`integuru/util/LLM.py`**
  - Import `Ollama` from `ollama`.
  - Add `_ollama_model` attribute to `LLMSingleton` class.
  - Modify `get_instance` method to support Ollama model.
  - Add `get_ollama_instance` method to return an Ollama instance.

* **`integuru/__main__.py`**
  - Update `model` option help text to include Ollama.

* **`README.md`**
  - Update documentation to reflect Ollama support.

README.md

@@ -62,6 +62,8 @@ Let's assume we want to download utility bills:
 
    **Recommended to use gpt-4o as the model for graph generation as it supports function calling. Integuru will automatically switch to o1-preview for code generation if available in the user's OpenAI account.** ⚠️ **Note: o1-preview does not support function calls.**
 
+   **Ollama support is now available! You can use the Ollama model by specifying `--model ollama` in the command.**
+
 ## Usage
 
 After setting up the project, you can use Integuru to analyze and reverse-engineer API requests for external platforms. Simply provide the appropriate .har file and a prompt describing the action that you want to trigger.

integuru/__main__.py

@@ -11,7 +11,7 @@
 
     @click.command()
     @click.option(
-        "--model", default="gpt-4o", help="The LLM model to use (default is gpt-4o)"
+        "--model", default="gpt-4o", help="The LLM model to use (default is gpt-4o, supports ollama)"
     )
     @click.option("--prompt", required=True, help="The prompt for the model")
     @click.option(

integuru/util/LLM.py

@@ -1,17 +1,22 @@
 from langchain_openai import ChatOpenAI
+from ollama import Ollama
 
 class LLMSingleton:
     _instance = None
     _default_model = "gpt-4o"  
     _alternate_model = "o1-preview"
+    _ollama_model = "ollama"
 
     @classmethod
     def get_instance(cls, model: str = None):
         if model is None:
             model = cls._default_model
 
-        if cls._instance is None:
-            cls._instance = ChatOpenAI(model=model, temperature=1)
+        if cls._instance is None or cls._instance.model != model:
+            if model == cls._ollama_model:
+                cls._instance = Ollama(model=model)
+            else:
+                cls._instance = ChatOpenAI(model=model, temperature=1)
         return cls._instance
 
     @classmethod
@@ -34,5 +39,10 @@ def switch_to_alternate_model(cls):
 
         return cls._instance
 
-llm = LLMSingleton()
+    @classmethod
+    def get_ollama_instance(cls):
+        """Returns an Ollama instance"""
+        cls._instance = Ollama(model=cls._ollama_model)
+        return cls._instance
 
+llm = LLMSingleton()