Notebook to PDF converting forces markdown title numbering

## Description

Converting a notebook with titles in the markdown sections to PDF adds odd numbering to **all** titles.
The quickest way to disable this feature is to append `{.unnumbered}` / `{-}` attribute to the title row, but they are visible in the notebook.

All other methods, such as using custom LaTeX templates, are painful and do not work.

## Reproduce

1. Create a new Notebook
2. Add two markdown sections
3. Create simple titles
4. Export to PDF via _File / Save and Export Notebook As / PDF_
5. The PDF contains titles with the numbering

## Expected behavior

Numbers must be where I put them.

## Context

- Operating System and version: Linux Ubuntu 24.04
- Browser and version: Latest Chrome and Firefox
- Jupyter Notebook version:
```bash
jupyter --version
Selected Jupyter core packages...
IPython : 9.5.0
ipykernel : 6.30.1
ipywidgets : not installed
jupyter_client : 8.6.3
jupyter_core : 5.8.1
jupyter_server : 2.17.0
jupyterlab : 4.4.7
nbclient : 0.10.2
nbconvert : 7.16.6
nbformat : 5.10.4
notebook : 7.4.5
qtconsole : not installed
traitlets : 5.14.3
```
Tested in the clean Jupyter installation in the dedicated venv.

Start by reproducing the export through File / Save and Export Notebook As / PDF in the reported clean Jupyter installation, then inspect the notebook PDF export path and its LaTeX handling. Done means headings retain only the numbers entered by the author without requiring visible {.unnumbered} or {-} attributes.

Written by the indexing model from the issue text.