fix: use input-length phases in parallel odd-even transposition sort

[mohammmad357]

Describe your change

• Add an algorithm?
• Fix a bug or typo in an existing algorithm?
• Add or change doctests? -- Note: Please avoid changing both code and tests in a single pull request.
• Documentation change?

The parallel odd-even transposition worker always performs ten phases. A
reverse-ordered eleven-item list therefore returns
[1, 0, 3, 2, 5, 4, 7, 6, 9, 8, 10] instead of list(range(11)).
Pass the input length to every worker and perform that many phases, matching
the existing comment and the odd-even sorting network's bound.

Refs #10925 (closed). @gyan313 previously identified and proposed this phase
count approach in closed, unmerged #10976; credit for that earlier observation
and approach belongs to them. This patch only carries the phase count through
the existing worker arguments. Empty/singleton input handling and typing
refactors remain outside this diff.

I checked the open parallel-sort PRs #15493, #15513, #15531 and #15538:
their patches retain the ten-phase loop. This addresses a separate correctness
case, although changes to the same file may conflict if another PR lands first.
It does not resolve the whole comparable-types umbrella #15234.

Validation

• Upstream base: 215c1f0710f610aaf5ebaeab04edacec3f5308cd, branch
  fix-parallel-sort-phase-count; exactly one algorithm file changed.
• Reproduced the eleven-item failure before the fix on Windows 10 / Python 3.13.7.
• After the fix, eight independent comparisons with Python sorted passed:
  reverse inputs of lengths 2, 3, 10, 11 and 12, twelve strings, fifteen seeded
  duplicate/negative integers and twenty seeded floats. All returned the same
  input list object. These checks use a standalone spawn-safe harness.
• All ten existing doctests passed; doctests are unchanged in this PR.
• git diff --check passed.
• Local Ruff/pre-commit could not run: installing their tools from PyPI timed
  out on files.pythonhosted.org; python -m pre_commit run --all-files
  reports No module named pre_commit. Python 3.15t was not tested locally.
  Leaving this as a draft until upstream CI and maintainer review are available;
  local correctness is not claimed to replace those checks.

Prepared with AI coding assistance; no outside user feedback or ratings are
claimed. The earlier contributor is explicitly credited above.

Checklist

• I have read CONTRIBUTING.md.
• This pull request is all my own work -- I have not plagiarized.
• I know that pull requests will not be merged if they fail the automated tests.
• This PR only changes one algorithm file. To ease review, please open separate PRs for separate algorithms.
• All new Python files are placed inside an existing directory. (No new files.)
• All filenames are in all lowercase characters with no spaces or dashes.
• All functions and variable names follow Python naming conventions.
• All function parameters and return values are annotated with Python type hints.
• All functions have doctests that pass the automated testing.
• All new algorithms include at least one URL that points to Wikipedia or another similar explanation.
• If this pull request resolves one or more open issues, then the description above includes the issue numbers with a closing keyword.

Unchecked items: this builds on the credited earlier approach; existing
unannotated parameters and worker doctest coverage are preserved; no new
algorithm or complete open issue is resolved. The newly added phases
parameter has an int annotation.