Annotate Vision-Language models with metadata attributes (~166 models)

[ooples]

Context

Sub-issue of #958 (Annotate all model classes with metadata attributes)

Scope

Annotate all concrete model classes in src/VisionLanguage/ with the 6 model metadata attributes.

Subdirectories (~166 models)

• InstructionTuned/ (~46 models)
• Encoders/ (~26 models)
• Generative/ (~13 models)
• Document/ (~12 models)
• Grounding/ (~11 models)
• Foundational/ (~9 models)
• VideoLanguage/ (~9 models)
• Unified/ (~8 models)
• Remaining subdirectories

Acceptance Criteria

• Every concrete model class has all required attributes
• dotnet build src/AiDotNet.csproj --framework net10.0 -c Release passes with 0 errors
• No model class in src/VisionLanguage/ is missed

[ooples]

All 168 vision-language model classes annotated with metadata attributes.

[ooples]

Audit Results — Vision-Language models: 100% COMPLETE

Coverage Summary (170 concrete model classes annotated)

Subdirectory	Models	Annotated	Status
InstructionTuned/	46	46	✅
Encoders/	26	26	✅
Generative/	16	16	✅
Document/	12	12	✅
Grounding/	11	11	✅
Foundational/	9	9	✅
VideoLanguage/	9	9	✅
Unified/	8	8	✅
Robotics/	7	7	✅
Reasoning/	6	6	✅
ThreeD/	6	6	✅
Medical/	5	5	✅
Editing/	3	3	✅
Proprietary/	3	3	✅
Total	170	170	All done

Notes

• Interfaces/ directory contains only interface definitions (17 files), not model classes — excluded
• All acceptance criteria met

Completed in PR #965. This issue can be closed when that PR merges.

Closes #968