image/tree: Chips to represent "in use"

[vvoland]

Replace In Use column with Extra (needs a better name though), that uses "chips" / "badges" to show whether the image is in use.

`docker images --tree` now shows metadata badges

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 2.84360% with 205 lines in your changes missing coverage. Please review.

Project coverage is 59.26%. Comparing base (17c5fe6) to head (c950d48).
Report is 90 commits behind head on master.

❌ Your patch status has failed because the patch coverage (2.84%) is below the target coverage (50.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@ Coverage Diff @@## master #5744 +/- ##
==========================================
- Coverage 59.42% 59.26% -0.17% 
==========================================
Files 347 353 +6 Lines 29402 29618 +216 ==========================================
+ Hits 17472 17552 +80 - Misses 10958 11086 +128 - Partials 972 980 +8 

[laurazard]

I really like this! I wonder if INFO would be better for that header, or maybe no header at all?

[vvoland]

Changed to Info, WDYT?

[tianon]

I'll echo what I said in the maintainer meeting yesterday: I don't mind the presentation, I like the idea of compressing the visible data where we can, but I do wonder whether we have enough users that care (or even should care) about whether or not an image has an attestation to display it by default. I'm all for the data being available on the CLI for custom --format strings so that tiny subset of users who do care about this data can add it, but adding it by default seems like a mistake (especially when it's such a lossy bit of data all by itself -- what kind of attestation? is this an SBOM? is this a provenance doc? is it a signature?).

"In Use", however, is something that I think is relevant to every user 👍

[vvoland]

Removed attested

[koitsu]

How does one go about disabling this unnecessary feature? More specifically, disabling the waste-a-line-of-terminal-output-and-expand-to-full-width-of-terminal line that appears at the top of output?

I have dug through c950d48 and I do not see a way to inhibit this behaviour (other than not using a pseudo TTY).