[#99] Tier 3: cover SIM_4 RUN_JEOD1x_compatible (time SIM_4 2/2)

[simnaut]

Summary

Closes the SIM_4_common_usage item of issue #99 Bucket B: the
environment/time SIM_4 verification sim has two RUNs and only
RUN_JEOD2x was covered. This adds RUN_JEOD1x_compatible (1/2 → 2/2).

Both RUNs share the same epoch (1998-12-31 00:00 UTC), 60 s cadence, and
span (t = 86460 s, crossing the 1999-01-01 leap second). They differ only
in the UTC/UT1 convention set in input.py:

• RUN_JEOD2x (already covered): default true_utc / true_ut1. UTC and
  UT1 track the leap-second / EOP tables, so UTC TJT jumps by 1 s at the
  boundary.
• RUN_JEOD1x_compatible (this PR): time_utc.true_utc = False,
  time_ut1.true_ut1 = False. Per JEOD time_converter_tai_utc.cc /
  time_converter_tai_ut1.cc, the true_* flags gate the table updates
  inside convert_a_to_b; with the flag false, the TAI−UTC and UT1−TAI
  offsets are frozen at their initialization values and never updated —
  the legacy JEOD 1.x convention. UTC/UT1 stay smooth constant offsets from
  TAI with no leap jump.

Reproduction (no new capability)

Our SimulationTime already supports the frozen-offset convention:
SimulationTime::new freezes the TAI−UTC offset, and set_ut1_tai_offset
freezes UT1−TAI (JEOD's override_data_table = true mode). The test computes
both offsets once at the epoch from our own LeapSecondTable
(tai_utc_at_tai_tjt → 31 s) and EopTable (ut1_minus_tai_seconds), then
holds them constant for the run (no live EOP table installed). The CSV's t=0
UT1 offset is used only as a one-time epoch sanity check, never fed back as a
per-step override — computational independence preserved.

Validated time scales (tier3_time_v4_jeod1x)

All logged JEOD columns over 1442 points:

scale	observed max err	tolerance
TAI TJT	7.9e-7 s	1e-6 s
UTC TJT	7.9e-7 s	1e-6 s
UT1 TJT	7.9e-7 s	1e-6 s
TAI seconds (elapsed)	0	1e-9 s

Residuals are JEOD calendar→TJT rounding noise at the epoch (sub-μs),
consistent with the existing RUN_JEOD2x bounds.

Changes

• trick/generate_references.sh: add RUN_JEOD1x_compatible to the SIM_4 group.
• New fixture time_v4_jeod1x_time_v4.csv (regenerated via cargo xtask regenerate-tier3).
• tier3_time_v4_jeod1x in tier3_sim_time_docker.rs (the established home for
  the time-scale sims; not a Simulation::step() trajectory).

Verification

• cargo nextest run -p astrodyn_verif_jeod -E 'binary(tier3_sim_time_docker)' — 7/7 pass
• cargo fmt --check — clean
• cargo clippy -p astrodyn_verif_jeod --tests -- -D warnings — clean

🤖 Generated with Claude Code

[Copilot]

Pull request overview

Adds Tier 3 coverage for JEOD SIM_4 RUN_JEOD1x_compatible, validating the legacy “frozen offset” UTC/UT1 convention (JEOD 1.x behavior) across the 1999-01-01 leap-second boundary, closing the remaining SIM_4_common_usage gap from issue #99.

Changes:

• Extend the SIM_4 reference-generation group to include RUN_JEOD1x_compatible.
• Add a new Tier 3 Rust test (tier3_time_v4_jeod1x) that validates frozen UTC (TAI−UTC held constant) and frozen UT1 (UT1−TAI held constant).
• Add the regenerated reference CSV fixture for the new RUN.

Reviewed changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated 3 comments.

File	Description
trick/generate_references.sh	Adds RUN_JEOD1x_compatible to the SIM_4 group and documents the two UTC/UT1 conventions.
crates/astrodyn_verif_jeod/tests/tier3_sim_time_docker.rs	Adds tier3_time_v4_jeod1x to cross-validate the frozen-offset convention against JEOD output.
crates/astrodyn_verif_jeod/test_data/time_v4_jeod1x_time_v4.csv	New JEOD reference CSV for RUN_JEOD1x_compatible.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.