test: add property and mutation coverage for core invariants

[mldangelo-oai]

Summary

Add deterministic, behavior-focused tests for the SDK's core invariants and a small mutation-testing trial.

Part 2 of 4 in the test-quality stack. #504 has merged; this PR is now based on main.

Changes

• Extract the pure cost calculations from filesystem-backed cost tracking while preserving the existing exports.
• Add seeded property tests for token costs, error handling, progress events, finding identity, sealed artifacts, and safe artifact paths.
• Test SQLite migration rollback, idempotence, and integrity against a real database.
• Strengthen malformed-event and Markdown-boundary cases, and make generated filenames portable to Windows.
• Add pinned development-only fast-check and Stryker dependencies. Keep mutation testing scoped to three pure modules.
• Preserve the new live-session events when merging the refreshed parent.

Testing

• Refreshed head eb839e7b, Bun 1.3.14, seed 1: 115 focused cost, error, progress, contract, publication, and comparison tests passed.
• Independent pure-module rerun, seed 12345: 28 passed, 0 failed.
• Type checking, formatting, and git diff --check: passed.
• Initial mutation baseline: 487 killed, 18 survived, 0 timed out; 96.44% score. This refresh does not introduce a mutation-score gate.
• Three fresh native reviews and independent verification of eb839e7b: no findings.
• The prior tree-equivalent head e4b3ef94 hit an unchanged Windows subprocess test and then a pre-existing replacement-lock test in its CI retry. The lock correction is handled separately in fix: preserve exact multiscan lock identities #563. CI will rerun on the new main-based head.

Risk and rollout

The production change is a pure cost-module extraction with the existing public exports retained. The package allowlist includes the new emitted module. Property and mutation tooling are development-only. The refreshed parent preserves default test randomization and the Windows timeout allowance. When replaying a shrink path, select the failing file and test name. Mutation testing remains diagnostic until repeated runs establish a useful baseline.

Public disclosure review

• No customer, partner, prospect, or user identities, data, or identifying details are included.
• No credentials, personal data, private source, scan findings, or nonpublic links or tickets are included.
• I reviewed the branch name, title, description, commits, changes, comments, logs, screenshots, attachments, and links for public disclosure.

[mldangelo-oai]

@codex review

Please review the current head e038344e35131ad0f6430df7786a18bf62b9aa7d.

[chatgpt-codex-connector]

Security review completed. No security issues were found in this pull request.

Reviewed commit: e038344e35

ℹ️ About Codex security reviews in GitHub

This is an experimental Codex feature. Security reviews are triggered when:

• You comment "@codex security review"
• A regular code review gets triggered (for example, "@codex review" or when a PR is opened), and you’re opted in so security review runs alongside code review

Once complete, Codex will leave suggestions, or a comment if no findings are found.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. More of your lovely PRs please.

Reviewed commit: e038344e35

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[mldangelo-oai]

@codex review

Please review the refreshed head 40b23460a1012505d4c41176a336b7074839f90b. This additive merge preserves the current parent, the 120-second Windows allowance, and the upstream live-session events. Normal and randomized SDK tests, packed-package checks, and the fresh native review gate passed.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Another round soon, please!

Reviewed commit: 40b23460a1

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Security review completed. No security issues were found in this pull request.

Reviewed commit: 40b23460a1

ℹ️ About Codex security reviews in GitHub

This is an experimental Codex feature. Security reviews are triggered when:

• You comment "@codex security review"
• A regular code review gets triggered (for example, "@codex review" or when a PR is opened), and you’re opted in so security review runs alongside code review

Once complete, Codex will leave suggestions, or a comment if no findings are found.

[mldangelo-oai]

@codex review

Please review the additive refresh at e4b3ef94bc73749956952a5181d6cdea6e2bb029 against the updated parent. The existing cost exports, package boundary, seeded properties, and diagnostic-only mutation setup are preserved.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Swish!

Reviewed commit: e4b3ef94bc

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[mldangelo-oai]

@codex security review

Please review the exact refreshed head e4b3ef94bc73749956952a5181d6cdea6e2bb029, including the pure cost-module extraction, artifact-path properties, and package allowlist.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Keep it up!

Reviewed commit: e4b3ef94bc

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

The base branch was changed.

[mldangelo-oai]

@codex review

Please review the main-based head eb839e7b10186fcf4c91d1efe99b15945e8b0271. #504 has merged. This additive restack has the same Git tree as e4b3ef94; the 115 focused tests and three fresh native reviews with independent verification passed again. The separately discovered, pre-existing lock-identity correction is in #563.

[mldangelo-oai]

@codex security review

Please review exact main-based head eb839e7b10186fcf4c91d1efe99b15945e8b0271, including the cost-module extraction, package allowlist, and artifact-path regression coverage.

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. 👍

Reviewed commit: eb839e7b10

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".