Support defining a region for read, show, save, snap for visual automation

[kensoh]

Following up on #457, where transparency (0% opacity) in images is supported in visual automation for advance use cases, raising an issue to create a new PR that allows defining regions for steps read, show, save, snap. This allows users to define a square/rectangular region on the screen to perform OCR for text extraction, and saving screenshot of that region.

[kensoh]

Above commit implements this feature. Prior to packaged release, feature available from cutting edge version here - https://github.com/kelaberetiv/TagUI#set-up

Below is the updated section in visual automation to be updated to readme and tutorial on next packaged release. Adding Also for these steps, (x1,y1)-(x2,y2) can be used as the identifier to define the region of interest on the screen to perform OCR or capture snapshot.

Transparency (0% opacity) is supported in .png images, for eg using an image of an UI element with transparent background to enable clicking on an UI element that appears on different backgrounds on different occasions.

Another example is an image of the window or frame (PDF viewer, MS Word, textbox etc) with the center content of the image set as transparent. This allows using read, show, save, snap steps to perform OCR and save snapshots for application windows, containers, frames, textboxes with varying content. Also for these steps, (x1,y1)-(x2,y2) can be used as the identifier to define the region of interest on the screen to perform OCR or capture snapshot.