smoldocling does not enrich images with captions

### Bug
When I analyze a PDF file, I would expect that images in this file will get enriched with a description as analyzed by a proper visual LLM model. But this does not happen.

### Steps to reproduce
run

```
docling --vlm-model smoldocling --image-export-mode placeholder --pipeline vlm --enrich-picture-description --num-threads 16 test.pdf
```

test.pdf is attached. The result test.md is also attached.

### Docling version
Docling version: 2.28.4
Docling Core version: 2.25.0
Docling IBM Models version: 3.4.1
Docling Parse version: 4.0.1
Python: cpython-313 (3.13.2)
Platform: Linux-6.13.9-200.fc41.x86_64-x86_64-with-glibc2.40

### Python version
3.13.2

[test.md](https://github.com/user-attachments/files/19684450/test.md)
[test.pdf](https://github.com/user-attachments/files/19684449/test.pdf)