Delete old checkpoints in stateDB automatically

Introduce a constructor parameter for the StateDB called `keep`, which tells
how many checkpoints to keep.  When doing a new checkpoint, if the number of
existing checkpoints exeeds `keep`, the oldest ones will be deleted.

batchbuilder/batchbuilder.go

@@ -29,7 +29,8 @@ type ConfigBatch struct {
 // NewBatchBuilder constructs a new BatchBuilder, and executes the bb.Reset
 // method
 func NewBatchBuilder(dbpath string, synchronizerStateDB *statedb.StateDB, configCircuits []ConfigCircuit, batchNum common.BatchNum, nLevels uint64) (*BatchBuilder, error) {
-	localStateDB, err := statedb.NewLocalStateDB(dbpath, synchronizerStateDB, statedb.TypeBatchBuilder, int(nLevels))
+	localStateDB, err := statedb.NewLocalStateDB(dbpath, 128, synchronizerStateDB,
+		statedb.TypeBatchBuilder, int(nLevels))
 	if err != nil {
 		return nil, tracerr.Wrap(err)
 	}

batchbuilder/batchbuilder_test.go

@@ -16,7 +16,7 @@ func TestBatchBuilder(t *testing.T) {
 	defer assert.Nil(t, os.RemoveAll(dir))
 
 	chainID := uint16(0)
-	synchDB, err := statedb.NewStateDB(dir, statedb.TypeBatchBuilder, 0, chainID)
+	synchDB, err := statedb.NewStateDB(dir, 128, statedb.TypeBatchBuilder, 0, chainID)
 	assert.Nil(t, err)
 
 	bbDir, err := ioutil.TempDir("", "tmpBatchBuilderDB")