This pull request mainly focuses on the integration of a new cognitive service module for computer vision (CV) and the refactoring of the VideoAnalyzer feature in the AIHub. The most significant changes include the addition of a new CV module in the infrastructure, modifications to the VideoAnalyzerController in the AIHub, and updates to the video player in the AIHub's user interface.

Infrastructure changes:

• infra/main.tf: Added a new variable cv_name to the locals block and a new module Computer Vision that uses the cv_name variable. [1] [2]
• infra/modules/cv/main.tf and infra/modules/cv/variables.tf: Created a new cognitive service account for computer vision. [1] [2]
• infra/modules/st/main.tf: Added a new storage container image-analyzer for the video analyzer feature.
• infra/variables.tf: Added a new variable cv_name with a default value.

AIHub changes:

• src/AIHub/Controllers/VideoAnalyzerController.cs: Refactored the DenseCaptionVideo method to use a new model for the cognitive service, improved code readability by using constants, and updated the payload structure for the chat API call. [1] [2] [3] [4]
• src/AIHub/Views/VideoAnalyzer/VideoAnalyzer.cshtml: Replaced the iframe for video display with a video tag.
• src/AIHub/appsettings.template.json: Updated the configuration for the VideoAnalyzer feature.

UI changes:

• src/AIHub/wwwroot/js/ui/videoanalyzer.js: Updated the video source setting method to support the new video tag in the user interface.