Follow-up to #2721. It fixes how the remaining tests behave and changes no production code.
## CI
- Both `dotnet test` steps pass `--filter "Category!=Manual"`. Before this, Manual-tagged tests ran on CI anyway.
- The one surviving movement-throttle latency test is the only coverage of burst forgiveness over a full, wrapped history. It loses its Manual tag, so it runs on CI.
## Tests that passed on CI without testing anything
- **Corpse and throwing tests.** They relied on tiledata for the weapon `Layer`, so on CI the equip silently failed and they asserted against empty lists. They now set `Layer` explicitly and assert the equip took effect. `TestThrown` also pins its own strength and range values, so a shard retuning SoulGlaive no longer breaks them.
- **Healthbar poison rows.** Server.Tests registers no poisons, so every row tested `null`. They now use real poison levels.
- **Account packet tests.** The mock accounts reported `Length` 0, so the character-list tests only ever encoded empty slots. They now encode the real characters.
- **Pathing and AI tests that only mean something on real Trammel/Felucca terrain** are now guarded with `TileDataRequirement.SkipIfMissing()`: Britain desk and clutter scenes, house footprints (no MultiData on CI) and fingerprints. Tests that are valid on flat blank terrain stay unguarded. (A synthetic tiledata fixture to run these on CI is a planned follow-up.)
- **Assertions that couldn't fail now check their claim.** Affected tests include the `DefaultWalker` cache use, cache on/off parity requiring a path, the recorder capability flags, the idle-pet gate, the monthly DST recurrence, the AdvancedSearch `Record.Exception` wrappers, `CanFitItem` with no branch-guarded assert, the Firewall callback, and `Import_NoMap`. The Geometry Parse asserts also now use the same input as their TryParse.
## Shared state
Every test that changes process-wide state now restores it, even when an assert fails:
- **The clock.** A new `TimerWheelScope` saves and restores `Core._tickCount`, `Core._now` and the timer wheel. `SimulatedClockScope` covers decay tests.
- **StepCache settings and caches**, via `StepCacheStateScope`.
- **Network state.** The ManualAllowlist, LoginAllowlist, firewall and BanChannel reporters are restored. Global packet handlers are now registered per test and removed afterwards, so they no longer stay registered for later tests.
- **Data and config tables.** `ArchiveJournal` statics, `TileData.ItemTable`, `ObjectPropertyList.Enabled`, `Core.Expansion` and the localization table are restored. One `MobileStatusVersion` restore was resetting a different expansion than the one the test changed.
- **Entities.** Tests with fixed serials now delete what they create. Fixed mobile serials moved to `World.NewMobile`, because deleting a hard-coded serial could unregister a live entity.
- **Collection race.** `SpiralScanTests` no longer clears the spawn cache from a separate collection that could run in parallel with the cache-manager tests.
- **Shared statics.** Classes reading shared static allowlists joined the sequential collection.
## Duplication
New shared helpers replace copy-pasted harness code:
- **Server.Tests:** `PacketAssert.Sends`, used by about 130 packet tests; `TestClientHarness`, `PersistencePipelineScope`, `NetTestHelpers`, `EntityCleanup` and one `TestMulti`.
- **UOContent.Tests:** `TimerWheelScope`, which also holds one wraparound-safe `RunFor`/`RunUntil`; `AiTestStubs`, which merges six walker copies and five identical creature stubs; `MobileTestHarness`, `CommandTestHarness` and `TempDirectory`.
Dead oracle packet classes are removed.
## Cost
| Test | Before | After |
|---|---|---|
| `VirtualSerialTests` | 1.57s | 2ms (starts near the wrap via reflection instead of drawing ~17.9M serials) |
| Event scheduler recurrence loop | 473ms | 1ms |
| CrowdSec drain | 517ms | 16ms (event-driven instead of a fixed 500ms wait) |
| AuthId tests | 248ms | 99ms (SHA2 except the two GameLogin tests) |
| Container cap tests | 5,000 items | 3,449 / 3,277 (one past each protocol cap) |
| Bulk removal | 100k children | 20k children |
| OPL hash distribution | 60k builds | 10k builds |
## Shard-tuning decoupling
- AI assertions read `ReacquireDelay` and `FollowMoveSpeed` instead of hard-coding production tuning.
- The guard friendly-fire test uses an `AI_Mage` stub instead of WhiteWyrm.
- Argon2 assertions pin the `argon2id` variant and `v=19` but read the cost parameters from the configured hasher.
## Tests
| Run | Server.Tests | UOContent.Tests |
|---|---|---|
| With client files | 812 pass | 1,123 pass |
| Without client files (CI simulation) | 800 pass, 12 skip | 1,021 pass, 102 skip |
- The UOContent fixture reads `MODERNUO_TEST_DATA_DIR`; Server.Tests reads `MODERNUO_CLIENT_PATH`.
- The 15 new UOContent skips are the map-dependent guards above.
- Total run time with client files is about 5s, down from about 8s.
## Summary
Stacked on #2676. Review that first; this PR's diff is against it.
Items players sell to NPC vendors go into the vendor's buyback (`ShopBuy`) pack. That pack was purged only when someone opened the buy menu, so a vendor that players sold to but never bought from grew without limit.
- A server operator reported 250k+ items piling up from players crafting and selling on a macro.
- Every one of those items was saved.
- Purging or deleting that vendor then sent one remove packet per item to every nearby client and disconnected them.
The behavior is inherited from RunUO.
- **`VendorBuybackPack`** replaces the plain backpack on `Layer.ShopBuy`.
- It opts in to `SkipsChildSerialization`, so buyback items are never saved.
- Its contents are private like any container's, so removing them reaches only the vendor (no client) and openers. Purges and deletes send nothing to bystanders.
- **Capacity.** The buyback holds at most `250 − buy-list entries` items, which is what the buy gump can show. The oldest items are evicted first. At capacity 0 the sold item is consumed and the seller is still paid.
- **Purge on the restock grid.**
- `Restock()` empties the buyback.
- A purge-only timer, armed on the first sale, covers vendors nobody buys from. It is due at the next restock boundary after now, so a vendor whose restock hasn't run for hours doesn't purge an item the moment it's sold.
- The timer never restocks, so stock doubling and halving is unchanged.
- The old per-item one-hour `LastMoved` check is gone.
- **Existing saves.** A legacy `ShopBuy` pack is replaced on the first tick after load. Thanks to #2676, deleting a pack of 250k+ items takes well under a second.
**Behavior changes for shard operators:**
- A buyback item lasts until the next restock boundary (at most `RestockDelay`), not one hour after its sale.
- A vendor that changes maps drops its buyback.
- A buy window that is open during a purge can show a row whose item is gone. Buying that row does nothing, and reopening shows the current list.
## Test plan
- [x] `VendorBuybackPackTests`: evicts oldest first; capacity 0 consumes the item; purge; a stale front entry can't spin eviction
- [x] `VendorBuybackTests`:
- purge due time stays on the restock grid, including a stale restock and a restock time in the future;
- partial stack sale: the split goes to buyback and the seller keeps the remainder;
- zero capacity consumes the item and still pays;
- restock purges and clears the timer, and the next sale re-arms it;
- deleting the vendor cancels the timer;
- a legacy 20k-item pack is replaced.
- [x] UOContent.Tests and Server.Tests green; schema generator clean
- [x] Manual in-game pass:
- sell and buy back, including a partial stack;
- over-capacity sale;
- purge with no buyer;
- `[restock`;
- vendor facet change;
- restart;
- legacy save with a large `ShopBuy` pack;
- mass selling with bystanders present;
- deleting a vendor with a full buyback.
Kamron Batman