fix: snapshot+tail WAL pin failure cleanup + true mid-executor epoch test

Finding 1: PlanRebuild snapshot+tail WAL pin failure now fail-closed
- InvalidateSession("wal_pin_failed_during_rebuild", StateNeedsRebuild)
- Snapshot pin released, session invalidated, no dangling state
- New test: E2_RebuildWALPinFailure_SessionCleaned

Finding 2: True mid-executor invalidation test
- Executor makes 2 successful progress steps (60, 70)
- Epoch bumps BETWEEN steps (real mid-execution)
- Third progress step fails — session invalidated
- Resources released via executor cancel
- New test: E2_EpochBump_AfterExecutorProgress

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

sw-block/engine/replication/driver.go

@@ -165,6 +165,7 @@ func (d *RecoveryDriver) PlanRebuild(replicaID string) (*RecoveryPlan, error) {
 		retPin, err := d.Storage.PinWALRetention(snapLSN)
 		if err != nil {
 			d.Storage.ReleaseSnapshot(snapPin)
+			s.InvalidateSession("wal_pin_failed_during_rebuild", StateNeedsRebuild)
 			d.Orchestrator.Log.Record(replicaID, plan.SessionID, "wal_pin_failed", err.Error())
 			return nil, fmt.Errorf("WAL retention pin failed: %w", err)
 		}