[Enesuygurs]

When reversed: true is set without start, the start number is taken from items.length. That count also includes items that are nested lists ({ ol: [...] } / { ul: [...] }). Those items get no marker and don't change the counter, so the list starts too high and never reaches 1.

{ reversed: true, ol: ['parent item 1', { ol: ['nested item 1', 'nested item 2'] }, 'parent item 2'] }

Before: 3., (nested list), 2.
After: 2., (nested list), 1.

The default start now counts only the items that actually get a marker, which matches how non-reversed lists already skip nested lists when numbering. Lists with an explicit start are unchanged.

I added a unit test in tests/unit/DocMeasure.spec.js and a CHANGELOG entry. npm run mocha gives 615 passing, and npm run lint is clean.

[liborm85]

Thanks