fix(uspto): parse proportional and missing CALS colwidth

[santhreal]

The USPTO table parser strips only pt/mm from colwidth then casts to int/float. A proportional width like colwidth="1*" (standard CALS) reaches float("1*") and raises ValueError, and a <colspec> with no colwidth reaches re.sub("pt", "", None) and raises TypeError. Both crashes are caught at the call site (XmlTable.parse() is wrapped in try/except Exception), so every affected table is silently dropped from the output document.

This extracts the leading numeric magnitude instead, defaulting to 1.0 when the width is absent, so proportional and unit-less tables parse with correct relative offsets rather than being lost.

Repro against the real backend:

from docling.backend.xml.uspto_backend import XmlTable

xml = (
    '<table><tgroup cols="2">'
    '<colspec colname="1" colwidth="1*"/><colspec colname="2" colwidth="2*"/>'
    '<tbody><row><entry>a</entry><entry>b</entry></row></tbody>'
    '</tgroup></table>'
)
XmlTable(xml).parse()  # before: ValueError: could not convert string to float: '1*'

Added test_table_colwidth_variants covering proportional, unit-less, and fixed-pt (control) widths.

[github-actions]

✅ DCO Check Passed

Thanks @santhreal, all your commits are properly signed off. 🎉

[mergify]

Merge Protections

🟢 Merge protection satisfied — ready to merge.

Show 1 satisfied protection

🟢 Enforce conventional commit

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)(?:\(.+\))?(!)?:

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[ceberam]

Good catch @santhreal ! I've just added a minor styling comment.

[ceberam]

@santhreal could you fix the DCO? (your last commit was not sign), as well as the conflict with tests/test_backend_patent_uspto.py?