feat(cli): add --image-alt-mode option for markdown image alt text

[s1v4-d]

Adds support for the new image_alt_mode parameter from docling-core, allowing users to choose how alt text is generated for images in markdown export: static, caption, or description.

Closes #2803

Checklist:

• Documentation has been updated, if necessary.
• Examples have been added, if necessary.
• Tests have been added, if necessary.

[github-actions]

✅ DCO Check Passed

Thanks @s1v4-d, all your commits are properly signed off. 🎉

[s1v4-d]

MyPy pre-commit check fails because this PR depends on docling-project/docling-core#463 being merged first. Once that's merged and docling-core is updated, mypy will pass.

[mergify]

Merge Protections

Your pull request matches the following merge protections and will not be merged until they are valid.

🟢 Enforce conventional commit

Wonderful, this rule succeeded.

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

• title ~= ^(fix|feat|docs|style|refactor|perf|test|build|ci|chore|revert)(?:\(.+\))?(!)?:

[dosubot]

Related Documentation

Checked 9 published document(s) in 1 knowledge base(s). No updates required.

^{How did I do? Any feedback?}  

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[vagenas]

Thanks @s1v4-d for the interesting idea.

First of all, some relevant context: the annotation & caption info (as available) is anyway shown by the serializer in the normal "flow", i.e. outside the link text.

Therefore, displaying it again as the link text (i.e. what goes in [here](...)) causes duplication, as shown in the screenshot [1] where I just dumped the output of a PR's test.

What would be a valid alternative is just parametrizing the text that's currently hard-coded to "Image" in [Image](...).
This would add some flexibility, while respective the current scheme & avoiding duplications.
Would you up for that?

[1]

[s1v4-d]

Thanks @s1v4-d for the interesting idea.

First of all, some relevant context: the annotation & caption info (as available) is anyway shown by the serializer in the normal "flow", i.e. outside the link text.

Therefore, displaying it again as the link text (i.e. what goes in [here](...)) causes duplication, as shown in the screenshot [1] where I just dumped the output of a PR's test.

What would be a valid alternative is just parametrizing the text that's currently hard-coded to "Image" in [Image](...). This would add some flexibility, while respective the current scheme & avoiding duplications. Would you up for that?

[1]

sure thing

[s1v4-d]

Run examples pipelines would pass once the pr on docling-core is merged and released.

[lcs-crr]

Would this feature only be available on the CLI?