Feature(util): Add a command to decode invisible watermarks

[lstein]

Summary

This adds a command, invoke-readwatermark that reads the invisible watermarks written to generated images when the "Enable Invisible Watermark" option is added. I realilzed we needed this during testing of PR #8871.

Related Issues / Discussions

QA Instructions

1. Install this PR with uv pip install... in order to create the command.
2. Generate an image or two with invisible watermarks.
3. Run invoke-readwatermark <path to image>... from the command line with images that should and should not contain watermarks.
4. Verify that command returns InvokeAI watermark for images that have the watermark and an empty string for those that don't.

Note that the watermarking package we use is very sensitive to the content of the image and seems to flip bits at random. Some combinations of model and VAE are particularly prone to generating garbled watermarks.

Merge Plan

Simple merge.

Checklist

• The PR has a short but descriptive title, suitable for a changelog
• Tests added / updated (if applicable)
• ❗Changes to a redux slice have a corresponding migration
• Documentation added / updated (if applicable)
• Updated What's New copy (if doing a release after this PR)