Add Computational Cost tab to GUI Evaluator

[sumitsaraswat362]

Fixes #560.

This PR adds a Computational Cost tab to the Streamlit GUI, achieving parity with the CLI's computational-cost behavior.

Changes

• Added a fourth tab: Computational Cost in app.py
• Implemented gui/computational_cost.py to route to the correct task
• Implemented cost calculation for Image Detection, Image Segmentation, and LiDAR Segmentation models.
• Exposes configurations for input shape (height/width or num_points/range/intensity), runs, and warm_up_runs.
• Displays metrics (Parameters, Model Size, Latency, FPS).
• Allows expanding the detailed results DataFrame and downloading it as a CSV.

Tested locally, all tabs render correctly and successfully pass arguments to the underlying Torch*Model.get_computational_cost() methods.

[sumitsaraswat362]

Hi team! I saw issue #560 was open and decided to tackle it. I've implemented the computational cost tab for Image Detection, Image Segmentation, and LiDAR Segmentation, keeping the UI style consistent with the rest of the evaluator. Please let me know if you'd like me to tweak the layout or add any additional metrics I'm happy to iterate on your feedback!