[trissim]

Configuration types registered during preparation hooks could reach registry READY before their source declarations were prepared. This prepares the current public/lazy dataclass pairs after all callable hooks and before READY, including when the callable catalog is already loaded. It reads the existing registry and source-preparation owners without constructing configuration values or evaluating default factories.

Pins ObjectState #9 plus its type-annotation correction #10 at 9c31e4b, and python-introspect #6 at 6f5ac0b. OpenHCS changes are limited to this preparation loop and its readiness regression test.

Validation on main d1103229 plus this change:

• The readiness regression fails on unmodified main and passes here. It checks hook-time registration, later registration with a loaded catalog, non-dataclass filtering, declaration order before READY, and no default-factory evaluation.
• Readiness, processing preparation, execution-server startup, and registry iteration tests: 21 passed, 5 skipped. The existing skips require two or four fork affinity slots; this run used CPU 3 only.
• Original unmodified R0 and R1: PASS, against the main-based commit 4971f4d4; R1 retained its 160-second budget. The registry is the only changed OpenHCS production module. Existing nominal ownership is retained; no new class or preparation cache is added.

No performance measurement on main is claimed. No canonical-signature preparation or other PR #394 runtime changes are included. Upstream python-introspect Python 3.10 retains three pre-existing recursive-projection failures; the new upstream tests pass.

Closes #513

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.