Consensus fail after updating validators

[cloud8little]

Describe the bug
A clear and concise description of what the bug is.

To Reproduce
Steps to reproduce the behavior:

1. Start a CN and normal node.
2. Register all standby committee as candidate and the original validator as candidate too.
3. Register another account test1.json as candidate and vote him with 20000000 votes.
4. After generating another two blocks. consensus node show error.

Expected behavior
It should not show error but show role=WatchOnly.

Screenshots

Platform:

• OS: Windows 10 x64
• Version master branch, Neo-VM

Root Cause Analysis:

        public void Reset(byte viewNumber)
        {
            if (viewNumber == 0)
            {
                Snapshot?.Dispose();
                Snapshot = Blockchain.Singleton.GetSnapshot();
                Block = new Block
                {
                    PrevHash = Snapshot.CurrentBlockHash,
                    Index = Snapshot.Height + 1,
                    NextConsensus = Blockchain.GetConsensusAddress(NativeContract.NEO.ComputeNextBlockValidators(Snapshot))
                };
                var pv = Validators;
                Validators = NativeContract.NEO.GetNextBlockValidators(Snapshot);
                if (_witnessSize == 0 || (pv != null && pv.Length != Validators.Length))
                {

It was caused by ComputeNextBlockValidators get the new update key, GetNextBlockValidators still get the original validators.