files deleted:
        storage/tianmu/core/rc_attr_typeinfo.h
        storage/tianmu/handler/tianmu_handler.cpp
        storage/tianmu/handler/tianmu_handler_com.cpp
        storage/tianmu/types/rc_data_types.cpp
        storage/tianmu/types/rc_num.cpp
        storage/tianmu/types/rc_num.h
        storage/tianmu/types/rc_value_object.cpp

Cause of the problem:
It is caused by writing back the field value when inserting the primary key and the problem of the assignment interface used.

Some restrictive features have been developed

Add the contact information module of the official

Modify the documentation in the Getting Started chapter

fix(tianmu):limit min signed int64 value to -9223372036854775807,The failure of mtr operation caused by missing the update logic

feat(tianmu):delta store,Modify the handler layer and modify the base class to support the delta layer

 bug: The error message displayed is inaccurate when you drop unique index or drop fulltext index. #1342,#1243

[summary]
1. change file(tianmu_attr_exeq_rs.cpp) format from dos 2 unix

2. rm incorrect between check, actually the v1 or v2 equal to boundary is illegal and it will cause empty result when condition
like `where col = PLUS_INF_64`, `where col = PLUS_INF_DBL`, `where col = MINUS_INF_DBL`:

```
common::RoughSetValue TianmuAttr::RoughCheckBetween(int pack, int64_t v1, int64_
                                                                // and then consider negation
   bool is_float = Type().IsFloat();
   auto const &dpn(get_dpn(pack));
-  if (!is_float && (v1 == common::PLUS_INF_64 || v2 == common::MINUS_INF_64)) {
-    res = common::RoughSetValue::RS_NONE;
-  } else if (is_float && (v1 == *(int64_t *)&common::PLUS_INF_DBL || v2 == *(int64_t *)&common::MINUS_INF_DBL)) {
-    res = common::RoughSetValue::RS_NONE;
-  } else if (!is_float && (v1 > dpn.max_i || v2 < dpn.min_i)) {
+  if (!is_float && (v1 > dpn.max_i || v2 < dpn.min_i)) {
```

Modify points:
1. Add (RENAME_INDEX) unsupported judgment conditions
2. Remove the redundant filtering logic in (log_event. cpp)

[summary]
In Tianmu engine:
For int64, -9223372036854775807 is reserved to indicate as `NULL_64` flag, so we limit
min value of int64 to -9223372036854775806.

For int32, -2147483648 is reserved to indicate as `NULL_32` flag, we limit min value of int32
to -2147483647.

[summary]
1. add support in load/export module
2. fix format with clint

[summary]
1. expand unsigned boundary from signed boundary to max valued.
2. code cover create/insert(delayed/no delayed mode)/update/delete/select.
3. functions and operators(agg and math calculation)

refs: https://github.com/stoneatom/stonedb/issues/1266
related issue: #1151

    1. The CACHE_ITEM type does not recurse by making a direct comparison
    2. When two referenced variables have the same address,
        they are considered to be the same value

Signed-off-by: NEl-even-11 <zhao_ziqian@sjtu.edu.cn>

Signed-off-by: NEl-even-11 <zhao_ziqian@sjtu.edu.cn>

Modify points:
1. Fix the error of judgment logic
2. Fix the value problem of (sql_mode) in the slave thread

bug(tianmu):Server crashed, when execute select val, val >= '2010-02-22 18:35:23' from (select * from timetest) as A;

Add test cases for NO_KEY_ERROR to test secondary index、unique index and trigger.

Add NO_KEY_ERROR to sql_mode.
Add judgement for secondary index、unique index、trigger、etc.

[summary]
1. support few column;
2. support too many column;
3. support the data of column is not completed,such as
3.1 such as start with the enclosed char, finished without enclosed char;
3.2 without line-termination string, but reach to the end of file;

After altering table add column, the data of the added column is NULL. When handling the
NULL data, the updating operation is not correct.

Cause of the problem:
There is a problem in the method of waiting for the master/slave synchronization to complete,
which leads to the data not being synchronized, and then proceed to the next step

Modify scheme:
Use the correct method to determine whether the slave database is synchronized

Because the records which was deleted is not filtered in fill_row().

feat(tianmu):improve the code quality and refactoring derived table optimize for tianmu(#1277, #1276, #1258)