fixed wrong task_name in send_task_inference_request inside _agentic_object_detection

[vishalk1995]

I was working with Vision Agent and encountered that even when the agentic_object_detection is called the results were produced by the text-to-object detection API.
For _agentic_object_detection the final URL call should be to "https://api.landing.ai/v1/tools/agentic-object-detection"

As per the this page:
https://va.landing.ai/demo/agentic-od

import requests url = "https://api.landing.ai/v1/tools/agentic-object-detection" files = { "image": open("{{path_to_image}}", "rb") } data = { "prompts": "{{prompt}}", "model": "agentic" } headers = { "Authorization": "Basic {{your_api_key}}" } response = requests.post(url, files=files, data=data, headers=headers) print(response.json())

[dillonalaird]

@vishalk1995 , thanks for showing interest in vision-agent! I have given an explanation for the reason we use a different endpoint in the issue here #382