fix(vendor): make the public-API sweep cover pcapkit.vendor (#1772)

[JarryShaw]

Please follow the guide below

• You will be asked some questions, please read them carefully and answer honestly

• Put an x into all the boxes [ ] relevant to your pull request (like that [x])

• Use Preview tab to see how your pull request will actually look like

• Searched for similar pull requests

• Followed the coding style (make pylint, make mypy, make isort)

• make test passes, and a test case covers the change. Narrow runs only, listed below.

• Added a changelog entry under docs/source/changelog/ and regenerated CHANGELOG.md, if the change is user-visible: N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

What is the purpose of your pull request?

Tick the commit type your subject line carries.

• fix — corrects a defect
• feat — adds a feature
• perf — changes performance, not behaviour
• refactor — changes neither behaviour nor performance
• test — tests only
• docs — documentation only
• ci — workflows or build tooling
• release — bumps the version or rolls up a distribution
• chore — anything else

---

Summary

tests/project/test_public_api.py skipped pcapkit.vendor via EXCLUDED_ROOTS, for want of requests and bs4, but the test extra has carried both since #507. Dropping the exclusion failed the package converse check twice. This PR fixes both failures:

• pcapkit.vendor.ipx leaked path, its sys.path temporary. It is now _path.
• pcapkit.vendor's Vendor, ModuleNotFoundError, stacklevel, VendorWarning and warn go in DELIBERATE_NON_EXPORTS, not __all__. pcapkit.vendor.__main__ uses that __all__ as its default crawler list, which is now commented.
• EXCLUDED_ROOTS is empty, and the walk pins pcapkit.vendor and pcapkit.vendor.ipx. Stale prose about the exclusion is dropped in both test files. Rebased onto main past docs(tests): correct stale claims in test docstrings and comments (#1103) #1764's edits to the same docstrings.
• Bite: each of the three reverts fails the sweep. Runs: tests/project: 930 passed, 1 skipped.

Closes #1772

[JarryShaw]

Verdict: GOOD TO GO at 7043ac50d. A Sonnet cross-review (author: Opus), after a prose fix round and a rebase onto main. Each of the three fixes fails the sweep when reverted. A rebase check found #1764's corrections kept and nothing dropped. tests/project passed 930.

[JarryShaw]

Conflicts with #1789 in tests/project/test_public_api.py and tests/vendor/test_crawler_reachability_unit.py; whichever merges second, I rebase.

[JarryShaw]

resolve conflicts

[JarryShaw]

On it: the rebase onto main past #1789 is running now, with a Sonnet check on the resolution. I'll push it when the check passes.

[JarryShaw]

Verdict: GOOD TO GO at b5f5fb1d8. Conflicts resolved: rebased onto main past #1789, and merges cleanly into the current main (c25c28c8c). A Sonnet check replayed the conflicts itself: #1789's corrections are kept, nothing is dropped, and the pcapkit/vendor hunks are byte-identical to the reviewed commit.