make relative-date regex digit quantifiers possessive

[alhudz]

The relative-date matchers are built from \d+[.,]?\d* and run across the whole input: PATTERN in freshness_date_parser via findall/sub, and the split_relative regex in dictionary.py via split. On a long run of digits the two adjacent digit quantifiers backtrack without ever changing the match, so cost grows superlinearly.

Repro: dateparser.parse('9' * 3200), or any long digit run reaching the relative parser.
Cause: the expression is matched at every position; splitting \d+ against \d* over a digit run backtracks quadratically in split, near-cubically for the freshness findall.
Fix: make the digit quantifiers possessive (\d++/\d*+). A digit run can only be followed by [.,] or the literal unit text, so giving digits back can never complete a match.

PATTERN.findall('9' * 3200) goes from ~23s to ~0.02s; match results are unchanged and the existing suite passes.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.11%. Comparing base (081d251) to head (3529261).
⚠️ Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1335      +/-   ##
==========================================
- Coverage   97.11%   97.11%   -0.01%     
==========================================
  Files         235      235              
  Lines        2912     2909       -3     
==========================================
- Hits         2828     2825       -3     
  Misses         84       84              

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[serhii73]

Thanks!