Separate cast plans from pointer reconstruction

[joshlf]

The numerical cast decisions are nested inside the generic pointer
projection, which makes them difficult to extract and prove independently.
Move them into a non-generic CastPlan with separate selection and metadata
conversion methods.

Keep CastParams<Src, Dst> as the witness connecting a plan to its
KnownLayout pair. Its associated constant preserves compile-time rejection
of unsupported casts. Pointer projection reads metadata, applies that plan
and reconstructs the destination pointer. Preserve the existing complete
size comparisons and unchecked arithmetic.

Agent-authored-by: AI agent acting on joshlf's behalf

---

• 　  Prove record construction against independent Rust references #3846
• 　  Prove selected cast plans preserve size without metadata overflow #3845
• 　  Prove CastPlan selection and unchecked metadata arithmetic #3844
• 　  Prove cast validation and exact-size metadata inference #3843
• 　  Prove allocation-size preparation and tail containment #3842
• 　  Prove CastUnsized's production layout acceptance gate #3841
• 　  Prove the numerical bounds used by split operations #3840
• 　  Prove advancement and padding-sequence decisions #3839
• 　  Prove DstLayout padding and field extension #3838
• 　  Prove unit and usize PointerMetadata operations #3837
• 　  Prove trailing sizes, padding and capacity including overflow #3836
• 　  Prove primitive DstLayout constructors #3835
• 　  Add an independent Rust reference for nested DST sizes #3834
• 　  Specify rounding encodings as alignment and phase #3833
• 　  Relate recursive record layout to normalized size formulas #3832
• 　  Prove normalized trailing-size and capacity formulas #3831
• 　  Reuse Aeneas rules for views and indexed loops #3830
• 　  Prove least padding and greatest aligned predecessors #3829
• 　  Test total and partial Aeneas contract semantics #3828
• 　  Verify Boolean and NonZeroUsize candidate predicates #3852
• 　  Verify the byte writers used by IntoBytes #3851
• 　  Prove U16 and U32 byte-order reads, writes and updates #3827
• 　  Prove guarded Boolean conversion and byte copying #3849
• 　  Reject unsupported execution effects before Aeneas lowering #3848
• 　  Check Rust specs against golden and live Aeneas models #3826
• 👉 Separate cast plans from pointer reconstruction #3825
• 　  Move cast_from into layout/cast_from.rs #3824
• 　  Share Rust layout and byte models in anneal/lean #3823
• 　  Update Anneal to Aeneas ca282ec and Lean 4.31 #3822

Latest Update: v8 — Compare vs v7

📚 Full Patch History

Links show the diff between the row version and the column version.

Version	v7	v6	v5	v4	v3	v2	v1	Base
v8	vs v7	vs v6	vs v5	vs v4	vs v3	vs v2	vs v1	vs Base
v7		vs v6	vs v5	vs v4	vs v3	vs v2	vs v1	vs Base
v6			vs v5	vs v4	vs v3	vs v2	vs v1	vs Base
v5				vs v4	vs v3	vs v2	vs v1	vs Base
v4					vs v3	vs v2	vs v1	vs Base
v3						vs v2	vs v1	vs Base
v2							vs v1	vs Base
v1								vs Base

⬇️ Download this PR

Branch

git fetch origin refs/heads/G5bbi6lfcow3sjrplwzweldyeyjggh3at && git checkout -b pr-G5bbi6lfcow3sjrplwzweldyeyjggh3at FETCH_HEAD

Checkout

git fetch origin refs/heads/G5bbi6lfcow3sjrplwzweldyeyjggh3at && git checkout FETCH_HEAD

Cherry Pick

git fetch origin refs/heads/G5bbi6lfcow3sjrplwzweldyeyjggh3at && git cherry-pick FETCH_HEAD

Pull

git pull origin refs/heads/G5bbi6lfcow3sjrplwzweldyeyjggh3at

Stacked PRs enabled by GHerrit.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-08T10:22:36.469381Z	ce12010	PR opened
🔒 Security Review	✅ Completed 2026-10-08T10:23:30.619702Z	ce12010	PR opened

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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 9.75610% with 74 lines in your changes missing coverage. Please review.
✅ Project coverage is 61.20%. Comparing base (9916d7c) to head (731cde3).

Files with missing lines	Patch %	Lines
zerocopy/src/layout/cast_from.rs	9.75%	74 Missing ⚠️

Additional details and impacted files

@@                          Coverage Diff                           @@
##           Glzjt2cwoexxyhj2f2lgnba524ajc2cqp    #3825       +/-   ##
======================================================================
- Coverage                              92.94%   61.20%   -31.75%     
======================================================================
  Files                                     17       20        +3     
  Lines                                   2466     3727     +1261     
======================================================================
- Hits                                    2292     2281       -11     
- Misses                                   174     1446     +1272     

☔ 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.