In this change list
1. validate HLL column when loading data, if data is invalid, this row
will be filtered.
2. seems as empty HLL when serializing invalid type of HLL data, with
this change, all ingested data will be valid.
3. seems as empty HLL when deserializing nullptr or invalid type of HLL data.
With this change, dirty data can be handled normally.
4. rename function empty_hll to hll_empty.
5. disable memtable_flush_execute_test because this will fails
sometimes. When tearing down, some thread is not joined, and they will
visit destroyed resource, which is invalid.
Now size of HyperLogLog struct is so large that it lead the rowset is
too small when ingesting data. In this CL, registers in HyperLogLog are
only created when it is needed. When ingesting data, it's normal case
that there are only few values in one HyperLogLog.
1. ix bugs in query layer.
2. remove some redundant code in BE
3. support specify multi helper node when starting FE
4. add proc 'cluster_load_statistic' to show load balance situation of Palo
