Task: Repair the equity routing

A position nested in a total return swap reaches its own trade's key instead of being written as a central equity row, so the routing and the stored type no longer disagree. The 13 unrouted codes stay unrouted.

The 17 routed codes are not re-pointed here. The catalogue's instrument column names the table that holds a code, and under the atomised shape the central row is what holds an instrument; a part reaches its central row by the key and is never a routing target. The re-pointing therefore lands in unit 5 with the header it names.

• A position nested in a total return swap is keyed by the trade that owns it and writes no equity central row.
• The catalogue routes no equity code to a table that does not hold it.
• The 13 codes with no importer keep an empty instrument.

(Implementation strategy. Written when work starts; key decisions are distilled into the parent story's * Decisions at close, but the plan itself stays — it is the historical record of what we did.)

Every number this task claims about a tree, with the command that produced it and the commit it was measured at. A number without both is not a measurement, because a clean-up record describes a tree and trunk moves under a branch: a consumer count, a gate's component count and a suite's size all change with no code changing, and the reader of a bare number cannot tell which tree it came from. Where the raw output matters, put it in an evidence/ file beside the task and name that file in the Command column.

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

A total return swap is now the composite the catalogue says it is. The import mapper no longer diverts a swap whose underlying is an equity position into the equity family, so no equity_position_instrument row is written for a trade whose type is a swap, and the routing and the stored type agree (E6).

The old path is removed rather than accommodated, per the operator: no compatibility branch, no migration, and the database was recreated. underlying_sub_trade_type, nested_sub_trade and the two fall-backs are deleted, and both reverse mappers lost outer_trade_type and the wrapper that rebuilt a swap around a position. Five files, 47 insertions and 112 deletions.

ores.ore.core.tests passes: 483711 assertions in 719 cases.

The corpus has no top-level position document, so the mapper test takes the sub-trade out of the swap fixture and maps it as the top-level trade it now is. That is a compromise worth naming: the position path is covered through a constructed top-level trade, not a documented one. A real top-level position fixture would be better and does not exist today.