Task: Bind ores.analytics entities to profiles; verify zero-diff regen

Survey every ores.analytics entity against the profile catalogue, rebind matches, regenerate, and require zero additional diff against the pre-rebind checked-in output.

• Every entity discovered via the analytics-cpp catalogue entry is surveyed against all Variability Profiles.
• Every matching entity is rebound and regenerated with zero additional diff versus its pre-rebind checked-in output.
• The known service_find_by_uuid/service_find_by_code/service_find_prefix gaps on pricing_model_config, pricing_model_product, pricing_model_product_parameter, and pricing_engine_type are resolved as part of reaching zero drift.
• No-match entities are recorded with their actual feature values, not silently dropped.
• Full local build and ctest green after regeneration.

The bind was already on main when this task was picked up. It landed on 2026-09-11 under a different task in a different environment, and neither this task nor its story was reconciled afterwards, so both still read BACKLOG. The survey below records what was found. What remained to do was the tidy, the survey record, and the reconciliation of the two task records.

The tidy removes the per-entity :has_tenant_id: the entity and its profile both set. Every other component bound so far carries the :profile: alone; business_unit is the one recorded exception. It keeps the structural properties its hierarchy needs, which no profile assigns, and it keeps :has_parent_id: explicit because the generator reads that feature at the entity root while the profile writes it into the sql namespace. Analytics holds no hierarchy entity, so neither case arises here.

The proof is the regeneration, not the reading. With the four models edited, regeneration changes no generated file: the diff is exactly the four model orgs and nothing else, and no untracked file appears. The pre-edit baseline was empty, so zero additional diff holds.

The build and the ctest suite are run after this commit by the user, who is driving builds in this session.

Survey of projects/ores.analytics/modeling, the modeling directory of the analytics-cpp catalogue entry. One entry covers the whole directory, so the survey runs once.

No entity is a no-match and none needs an exception. All four resolved to a profile, and both profiles used carry a Physical-space table, so the eventing integration tests are enabled for all four.

The composite component ores.analytics holds four parts. Only api, core and service are codegen surfaces that carry entities; quant is a hand-written quantitative library with no entity models, and nothing in it is codegen-eligible.

The three recorded starting-point wrinkles were already resolved before this task started, and regenerating at the head of this branch confirms it: service_find_prefix on pricing_engine_type, service_find_by_uuid on the other three, and service_find_by_code_column: name on pricing_model_config.

system_tenant_validation is a variability feature that no profile assigns, so pricing_engine_type keeps it explicitly. That is the correct shape, not a leftover.

Manual QA scenarios (scaffolded via compass add test_scenario, run through the QA Validation Runner panel) that verify this task. Link new ones here as they're created; the scenario doc itself links back via its "Verifies task" field.

The survey found the bind already done. All four analytics entities carried a :profile: on main when this task started, and the regeneration confirmed every one of them byte-identical to the checked-in tree.

What this task changed is small and deliberate. It removed the per-entity :has_tenant_id: true from the four models, because both profiles already fix that feature to true and the explicit line was redundant. It recorded the survey and its disposition per file. It reconciled the superseded ad hoc task, which stays DONE because it delivered the work through PR #1893.

The proof of the tidy is the regeneration. With the four models edited, regeneration moved no generated file. The diff was the four model orgs and nothing else, and no untracked file appeared. The pre-edit baseline was empty, so zero additional diff holds.

check_component_drift.py --component analytics-cpp reports no drift on the committed tree.

Acceptance:

• Survey of every entity in the analytics-cpp catalogue entry: met. The table in * Notes covers all six modeling files.
• Zero additional diff on regeneration: met, by the argument above.
• The three recorded service flag wrinkles: met. They were already resolved on main, and the regeneration confirms it.
• No-match entities recorded with their values: not applicable. No entity is a no-match, so nothing was dropped.
• Full local build and ctest green: the user runs both after this commit. This result does not claim a build or test outcome that this session did not observe.