Add Falcon Perception model support to inference-models

[hansent]

Integrate TII's Falcon Perception (600M param dense Transformer) for open-vocabulary object detection and instance segmentation from natural language text prompts. Wraps the falcon-perception==1.0.0 PyPI package using its BatchInferenceEngine for the simplest inference path.

• Two model classes: detection-only (OpenVocabularyObjectDetectionModel) and segmentation (InstanceSegmentationModel), both torch-only
• Registered in models_registry as "falcon-perception" architecture
• falcon-perception added as optional extra (conflicts with JP6/older CUDA extras due to torch>=2.11.0 + numpy>=2 requirements)
• Configuration constants for mask threshold, max tokens, image sizes
• 20 unit tests for post-processing helpers (bbox pairing, coordinate conversion, image format handling, edge cases)

What does this PR do?

Related Issue(s):

Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactoring (no functional changes)
• Other:

Testing

• I have tested this change locally
• I have added/updated tests for this change

Test details:

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code where necessary, particularly in hard-to-understand areas
• My changes generate no new warnings or errors
• I have updated the documentation accordingly (if applicable)

Additional Context