Compose same-plan Harass and Reinforcements with original reserve costs

docs/RULE_DECISIONS.md

@@ -758,7 +758,7 @@ These are unresolved in the cited records. Do not infer an answer from elapsed t
 
 | Boundary | Still unresolved | Existing record |
 | --- | --- | --- |
-| Reinforcements cost | The user question still asks whether to keep the card disabled, always transfer three reserve forces on every outcome, or waive only a sole traitor-call winner's cost; no answer is assumed. The bounded classic and Ecaz/Moritani prototype provisionally charges on all revealed outcomes, chooses normal reserves before elite and counts the transfer as battle losses. Those are product interpretations, not a ruling. Same-plan Harass/Stone Burner remain gated; opposing Harass is supported. Eligibility never depends on concealed matching Traitors. | [Source cost review](ECAZ_TREACHERY_RULES.md#14-september-reinforcements-cost-review) · [Bounded runtime](REINFORCEMENTS_RUNTIME.md) |
+| Reinforcements cost | The earlier all-outcome versus sole-Traitor-winner question remains unanswered. The bounded prototype provisionally charges on every revealed outcome, chooses normal reserves first and counts the transfer as battle losses. Under standing choose-recommended, same-plan Harass composes by paying that original reserve quote before returns; future returns cannot fund or change it. Those are product interpretations, not publisher rulings or a numerical user answer. Stone/Reinforcements and other module gates remain. Eligibility never depends on concealed matching Traitors. | [Source review](ECAZ_TREACHERY_RULES.md#14-september-reinforcements-cost-review) · [Combined policy](REINFORCEMENTS_RUNTIME.md#combined-utility-plan--16-october-2026) |
 | Harass / Stone Burner | The earlier question (keep gated or compare after withdrawal) remains unanswered. Under the standing 14 October choose-recommended instruction, the development prototype now provisionally compares undialed counters remaining after successful withdrawal. Opposing Traitor cancels the return; original fixed Occupy commitment stays. Public preflight considers possible fixed Ecaz withdrawal without reading hidden custody. Allocation ambiguity, Ix/Poison Tooth and other module boundaries remain. This is an implementation interpretation, not a publisher ruling or a user answer. | [Provisional timing](HARASS_WITHDRAW_RUNTIME.md#provisional-stone-burner-timing--16-october-2026) |
 | Recruits late accounting and rates | The earlier one/two-further-free question remains unanswered. The content-first prototype now preserves paid spice, deducts recorded actual free usage from the doubled rate and all returns from the total cap: two-free/one-paid Atreides can take two more free. This is visibly provisional, not a numerical user ruling. Unknown ledgers, pending repricing, late rate changes and same-turn replay remain guarded. | [Source alternatives](RECRUITS_RULES.md), [provisional runtime](RECRUITS_RUNTIME.md) |
 | Sapho remaining timing | First versus unspent Advanced Guild priority, partly completed combined turns, cyclic auction duration and later battle timing stay guarded. The existing pre-plan opportunity supports [early aggressor](SAPHO_AGGRESSOR.md). Prior questions remain unanswered. | [Source audit](JUICE_OF_SAPHO_SOURCE_UPDATE.md), [battle review](JUICE_OF_SAPHO_BATTLE_REVIEW.md) |