perf: evaluate each directional-sea phase and depth profile once

[SMI-Lab-Inha]

Summary

Speeds up the directional and multi-train sea kinematics (CD_Sea_Kinematics: wavetrain rows, spectral waves rows, WaveSpreading).

On the release-equivalent build, these kinematics take 80 % of examples/chain_two_train_sea.dat. For every node, step and component, the routine computed:

• the phase cosine twice: once for the surface elevation and again for the kinematics;
• two exponentials for the depth profile.

Changes:

• The elevation pass keeps its cosines for the kinematics pass, for up to 4096 components; any further ones are recomputed. They are computed in the same 256-component chunks as before.
• The depth profile uses EXP(-k (z + 2h)) = e2kd / EXP(k z) with the table's e2kd = EXP(-2 k h), as the JONSWAP component path CD_Component_Wave_Kinematics already does.
• The two-exponential form is kept wherever the shortcut would not be exact:
  • a component whose EXP(k z) or e2kd underflows;
  • a call at another depth than the synthesis. CD_SeaType now records its synthesis depth for this test.

Speed (IFX 2025.3.2 release-mirror build, one thread, median of 3, isolated runs):

Deck	Before	After
chain_two_train_sea	1.89 s	1.57 s (−17 %)
Decks without a directional sea (multibody, lazy-wave RAO, line break, buoy, static)	—	unchanged within noise

Results. The formulas are unchanged; the evaluation is reorganised.

• gfortran: the outputs of both directional-sea examples are bit-identical.
• IFX: the release compiler had computed the second-pass cosine through a fused sin/cos evaluation, which can differ from the first-pass cosine in the last bit.

Deck (IFX)	Channels differing	Maximum difference
chain_two_train_sea	none	0 (bit-identical)
chain_torsethaugen_spread	AnchTen1 only, isolated rows from row 1813	one unit in the 8th printed digit (1.2e-8 relative)

Tests

• New case_kinematics_reference in tests/test_wave_spectra.f90. It compares CD_Sea_Kinematics (elevation, velocity, acceleration, dynamic pressure) against a direct per-component sum of the original formulas, to 1e-11 relative. It covers four seas:

  1. 9 × 200 components, all cosines kept;
  2. 31 × 200 components, past the kept length;
  3. a short sea in 1000 m, where both exponentials underflow for many components (the test asserts the underflow is reached);
  4. a call at another depth than the synthesis.

  The sampled points are below the surface, near the seabed, and one above the surface.

• Full suite, Windows, gfortran 16.2, Release, slow tier included: 245/245 pass.

• pre-commit run --all-files passes.

Checklist

• pre-commit run --all-files passes
• ctest --test-dir build --output-on-failure passes
• python -m pytest tests python/tests passes (if Python or validation records changed) — not applicable
• VALIDATION.md updated if a validation result changed — no validation result changes
• CHANGELOG.md updated under [Unreleased] (heading added if missing) for user-visible changes
• Documentation in doc/ updated for changed inputs, options, or outputs — not applicable, no inputs or outputs changed
• Conventional Commit title (feat:, fix:, docs:, ...)
• Commits are signed off (git commit -s) under the
  Developer Certificate of Origin

[SMI-Lab-Inha]

@codex review

[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 2026-10-14T10:13:48.709546Z	07c0b9e	Manual request

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.

[read-the-docs-community]

Documentation build overview

📚 CableDyn | 🛠️ Build #35139256 | 📁 Comparing 07c0b9e against latest (f27c5e3)

  🔍 Preview build  

2 files changed

± changelog.html
± development.html

[chatgpt-codex-connector]

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

Reviewed commit: 07c0b9e5c9

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".