Improve tagger preview image formatting

[smith113-p]

Description

The primary change here is to render portrait cover images as portrait in tagger. Because Stash already renders the scraped image in portrait, this change makes it easier/possible to tell whether the image is the same or not.

Also makes minor formatting changes to improve how this looks particularly with the duration overlay:

• Standardizes preview image/video container dimensions (14rem x 10rem) with a subtle dark background.
• Formats and aligns the overlay duration badge cleanly on top of both preview cards and scraped scene images.

Includes the changes from #7193; I will rebase once that PR lands.

Related Issue

Fixes #7202

Testing

Manually checked how vertical and horizontal cover images look in tagger.

Screenshots

After:

Checklist

• I have read and understood the Contributing document.
• I have read and understood the AI Usage Policy document.
• I have made corresponding changes to the documentation (if applicable).

AI Usage Disclosure

• I have used AI tools to assist with this pull request, and I have disclosed the tools and how I used them below.

I used AI tools (Codex aka ChatGPT and Antigravity) to draft and refine the CSS formatting changes.

Additional Context

[DogmaDragon]

This PR used https://github.com/smith113-p/stash/tree/link branch as base, not develop.

[smith113-p]

Yes -- I was going to keep this as a draft until that PR lands. Also the PR description/title is not final.