fix: Ruin and Recreate index out of bound

[zepfred]

The score corruption occurs because the CH adds values to destination entities without recording a list change event beforehand. Let's assume the following example:

Before:
v1 -> c1 - c2 - c3
v2 -> c4

Ruined customers: 
c1 and c2

After:
v1 -> c3
v2 -> c4 - c1 - c2

Since only values from v1 are selected, the R&R move will record the before list change event only for v1, while erasing v2 as part of the rollback actions. We cannot predict which destination entities will be updated, which makes it difficult to record the necessary changes before the list is modified. The list R&R move is failing with the following:

2025-01-06 13:16:22,028 ERROR [org.acm.veh.res.VehicleRoutePlanResource] (pool-8-thread-1) Failed solving jobId (3ff26dab-05ae-4e47-af4d-53d671056ce9).: java.lang.IndexOutOfBoundsException: Index 8 out of bounds for length 2
        at java.base/jdk.internal.util.Preconditions.outOfBounds(Preconditions.java:100)
        at java.base/jdk.internal.util.Preconditions.outOfBoundsCheckIndex(Preconditions.java:106)
        at java.base/jdk.internal.util.Preconditions.checkIndex(Preconditions.java:302)
        at java.base/java.util.Objects.checkIndex(Objects.java:385)
        at java.base/java.util.ArrayList.remove(ArrayList.java:551)
        at ai.timefold.solver.core.impl.domain.variable.descriptor.ListVariableDescriptor.removeElement(ListVariableDescriptor.java:158)
        at ai.timefold.solver.core.impl.heuristic.selector.move.generic.list.ruin.ListRuinRecreateMove.doMoveOnGenuineVariables(ListRuinRecreateMove.java:67)
        at ai.timefold.solver.core.impl.heuristic.move.AbstractMove.doMoveOnly(AbstractMove.java:26)
        at ai.timefold.solver.core.impl.heuristic.move.LegacyMoveAdapter.execute(LegacyMoveAdapter.java:54)
        at ai.timefold.solver.core.impl.score.director.AbstractScoreDirector.doAndProcessMove(AbstractScoreDirector.java:265)
        at ai.timefold.solver.core.impl.localsearch.decider.LocalSearchDecider.doMove(LocalSearchDecider.java:123)
        at ai.timefold.solver.core.impl.localsearch.decider.LocalSearchDecider.decideNextStep(LocalSearchDecider.java:101)
        at ai.timefold.solver.core.impl.localsearch.DefaultLocalSearchPhase.solve(DefaultLocalSearchPhase.java:71)
        at ai.timefold.solver.core.impl.solver.AbstractSolver.runPhases(AbstractSolver.java:83)
        at ai.timefold.solver.core.impl.solver.DefaultSolver.solve(DefaultSolver.java:197)
        at ai.timefold.solver.core.impl.solver.DefaultSolverJob.call(DefaultSolverJob.java:125)
        at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:317)
        at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
        at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
        at java.base/java.lang.Thread.run(Thread.java:1583)
        Suppressed: java.lang.IllegalStateException: workingInitScore > 0 (1).
Maybe a custom move is removing more entities than were ever added?

                at ai.timefold.solver.core.impl.score.director.AbstractScoreDirector.assertInitScoreZeroOrLess(AbstractScoreDirector.java:255)
                at ai.timefold.solver.core.impl.score.director.AbstractScoreDirector.afterListVariableElementAssigned(AbstractScoreDirector.java:429)
                at ai.timefold.solver.core.impl.move.director.ListVariableBeforeUnassignmentAction.undo(ListVariableBeforeUnassignmentAction.java:12)
                at ai.timefold.solver.core.impl.move.director.VariableChangeRecordingScoreDirector.undoChanges(VariableChangeRecordingScoreDirector.java:64)
                at ai.timefold.solver.core.impl.move.director.EphemeralMoveDirector.close(EphemeralMoveDirector.java:47)
                at ai.timefold.solver.core.impl.score.director.AbstractScoreDirector.doAndProcessMove(AbstractScoreDirector.java:264)
                ... 10 more

[triceo]

@zepfred FYI we recently noticed a score corruption in R&R. Very well may be a symptom of the same problem. Please talk to @Christopher-Chianelli, he investigated it and may be on his way to fix it - more importantly, let's not have two people working on possibly the same thing.

[zepfred]

@zepfred FYI we recently noticed a score corruption in R&R. Very well may be a symptom of the same problem. Please talk to @Christopher-Chianelli, he investigated it and may be on his way to fix it - more importantly, let's not have two people working on possibly the same thing.

It is indeed a case of score corruption. Thank you for letting me know.

[zepfred]

Regression introduced by 30c97c5