feat: add support for FastSAM model with point, box and text prompts

[barhanc]

Description

Adds support for FastSAM model with required postprocessing for point, box and text (using already existing CLIP export) prompts. Also adds an example app to test these.

Since FastSAM uses YOLO instance segmentation backbone with some clever postprocessing to imitate Facebook's SAM (see https://docs.ultralytics.com/models/fast-sam/#model-architecture), we use the existing instance segmentation C++ implementation and add TS postprocessing to minimize code duplication.

Introduces a breaking change?

• Yes
• No

Type of change

• Bug fix (change which fixes an issue)
• New feature (change which adds functionality)
• Documentation update (improves or adds clarity to existing documentation)
• Other (chores, tests, code style improvements etc.)

Tested on

• iOS
• Android

Testing instructions

• Run the Computer Vision - Segment Anything app screen and test two available models there
• You can also run Computer Vision - Instance Segmentation app screen and test the two newly added models there.
• Check HF pages for the exported model https://huggingface.co/software-mansion/react-native-executorch-fast-sam

Screenshots

You can use following image for testing.

https://upload.wikimedia.org/wikipedia/commons/c/cd/Animal_diversity_October_2007.jpg

Related issues

Closes #555

Checklist

• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have updated the documentation accordingly
• My changes generate no new warnings

Additional notes