[JarryShaw]

• 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
• Added a changelog entry — 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?

• ci — changes to CI configuration files and scripts

---

Description of your pull request and other information

Closes #2115. CI's unittest-ordering matrix had no corekit leg, and util/run_unittest_leg.py corekit failed on main with 34 failures and 31 errors.

• The failures (already fixed on main, test-only 66a70edf4): bisection pinned the polluters to test_bitfield_unnamed_bits_1487_unit and test_decorators_path_1735_unit, which call reimport_once_per_class. Leaving the fresh pcapkit import in place is intended in this leg (test: three modules pass alone but fail together, and pytest reports it green #981): it is how the leg catches modules that hold a stale import. The enum tests did hold one: they bound EnumLookup/EnumKeyError/EnumValueError at module load, so the fix imports those names per test. With it, the corekit leg runs 903 tests with 0 failures.
• This PR: adds corekit to the matrix. integration stays out on purpose, as the workflow comment and run_unittest_leg.py record: is_unit_tier() rejects every file in it, so the leg would run nothing of its own, and the integration job already runs it under pytest. The stale comment and the "not legs" sentence in conventions/test-coverage.rst are updated.
• Test: test_unittest_ordering_shards now derives the expected legs from every tests/ subdirectory with a test_*.py, minus NOT_LEGS = {'integration'}. Against the old workflow it fails 2 checks, both naming corekit; it passes with the change.

[JarryShaw]

GOOD TO GO at 846986ebe (Sonnet review).

• Scope: only the workflow matrix, conventions/test-coverage.rst and the pin test change. The corekit leg has the same shape as the other plain legs. At about 233 s locally it is comparable to const, well inside the 20-minute step and 45-minute job caps, and it runs under cap_file_size.py.
• integration stays out: FIXTURE_TIER_DIRS makes is_unit_tier() reject every integration file, and test_integration_has_no_unit_tier_module pins it. No other tests/ subdirectory is uncovered, since protocols (rest) rglobs misc and schema.
• Pin test: fails 2 checks with the workflow reverted (['corekit'] != []) and 2 with cli deleted. Six other workflow-pin files pass.
• Not gating: required-checks does not need unittest-ordering (main-push only), so the new check name cannot block a merge.

[JarryShaw]

Ready to merge at 846986ebe.

• Review: the Sonnet review is GOOD TO GO.
• CI: every blocking check passed (Required checks passed SUCCESS). The 3.8–3.10 KTLO legs carry no warnings.
• Ordering: the leg this PR adds runs only on main. On a scratch tree, util/run_unittest_leg.py corekit already passes on main since 66a70edf4: 903 tests, 0 failures. So the new Plain unittest ordering (corekit) check should be green on its first push.

This closes #2115.

[github-actions]

Coverage: 91.54% (unit tier, Python 3.15, 846986ebe, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	2	95.16%
pcapkit/const	19599	880	2618	866	92.12%
pcapkit/corekit	2565	71	812	37	96.56%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	2939	23	1024	25	98.79%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17414	219	4574	185	98.14%
pcapkit/toolkit	869	36	298	10	95.20%
pcapkit/utilities	473	7	136	5	98.03%
pcapkit/vendor	4432	1988	1018	167	51.87%

Per-file detail: the coverage-html artifact of this run.