Catalog autonomous_execution refs in specs/sdl/references.md #1530

Problem

specs/sdl/references.md has no row for any reference field under behavior_specifications.*.autonomous_execution on dev a4357f9a. The v1 and v2 autonomous execution policies already carry references such as participant_implementation_ref, clock_ref, progression_policy_ref, temporal_constraint_refs and observation_boundary_ref (raes/participant_execution.py:89-98, :156-162). tools/check_sdl_catalog_parity.py passes, so its parity expectations do not cover these fields either.

#1467 (DSL-121) adds autonomous_execution.resource_budget.dimensions.*.tool_affordance_refs, which also gets no catalog row. Its body says:

The reference catalog in specs/sdl/references.md lists no autonomous_execution refs today, so the new tool_affordance_refs gets no catalog row; tools/check_sdl_catalog_parity.py passes.

#1469 adds rows for the aggregate behavior_specifications.*.resource_budget fields (specs/sdl/references.md:433 at e7df797f) but none for the v3 profile's fields.

Scope

• Add a catalog row, with target section, resolution stage and failure behavior, for every reference field under autonomous_execution, including the v3 resource_budget fields.
• Extend the catalog parity expectations so an uncatalogued autonomous_execution reference field fails tools/check_sdl_catalog_parity.py.

Done when

• specs/sdl/references.md has a row for every reference field under behavior_specifications.*.autonomous_execution.
• tools/check_sdl_catalog_parity.py fails when one of those rows is removed, and passes with all of them.

Related

• feat: support authored participant interaction budgets #1467 (noted it), feat: make interaction budgets a first-class participant behavior member #1469, Participant Interaction Budget Surface (DSL-121) #308 (DSL-121)