Support common table expression(CTE) in Nereids：
- Just implemented inline CTE, which means we will copy the logicalPlan of CTE everywhere it is referenced;
- If the name of CTE is the same as an existing table or view, we will choose CTE first;
`BackendServiceProxy.getInstance()` uses the round robin strategy to obtain the proxy,
so when the current RPC request is abnormal, the proxy removed by
`BackendServiceProxy.getInstance().removeProxy(...)` is not an abnormal proxy.
Proposed changes
1. function interfaces that can search the matched signature, say ComputeSignature. It's equal to the Function.CompareMode.
- IdenticalSignature: equal to Function.CompareMode.IS_IDENTICAL
- NullOrIdenticalSignature: equal to Function.CompareMode.IS_INDISTINGUISHABLE
- ImplicitlyCastableSignature: equal to Function.CompareMode.IS_SUPERTYPE_OF
- ExplicitlyCastableSignature: equal to Function.CompareMode.IS_NONSTRICT_SUPERTYPE_OF
3. generate lots of scalar functions
4. bug-fix: disassemble avg function compute wrong result because the wrong input type, the AggregateParam.inputTypesBeforeDissemble is use to save the origin input type and pass to backend to find the correct global aggregate function.
5. bug-fix: subquery with OneRowRelation will crash because wrong nullable property
Note:
1. currently no more unit test/regression test for the scalar functions, I will add the test until migrate aggregate functions for unified processing.
2. A known problem is can not invoke the variable length function, I will fix it later.
Refactor eliminate outer join #12985
Evaluate the expression with ConstantFoldRule. If the evaluation result is NULL or FALSE, then the elimination condition is satisfied.
after project, some Slot maybe project to another one. So we need to replace ExprId in DistributionSpecHash to the new one. if we do project other than Alias, We need to return DistributionSpecAny other than child's DistributionSpec.
This rule eliminate project that output set is same with its child. If the project is the root of plan, the elimination condition is project's output is exactly the same with its child.
The reason to add this rule is when we do join reorder in optimization, the root of plan after transformed maybe a Project and its output set is same with the root of plan before transformed. If we had a Project on the top of the root and its output set is same with the root of plan too. We will have two exactly same projects in memo. One of them is the parent of the other. After MergeProject, we will get a new Project exactly same like the child and need to add to parent's group. Then we trigger Merge Group. Since merge will produce a cycle, the merge will be denied and we will get a final plan with two consecutive projects.
## for example:
**BEFORE OPTIMIZATION**
```
LogicalProject1( projects=[c_custkey#0, c_name#1]) [GroupId#1]
+--LogicalJoin(type=LEFT_SEMI_JOIN) [GroupId#2]
|--LogicalProject(...)
| +--LogicalJoin(type=INNER_JOIN)
| ...
+--LogicalOlapScan(...)
```
**AFTER APPLY RULE: LOGICAL_SEMI_JOIN_LOGICAL_JOIN_TRANSPOSE_PROJECT**
```
LogicalProject1( projects=[c_custkey#0, c_name#1]) [GroupId#1]
+--LogicalProject2( projects=[c_custkey#0, c_name#1]) [GroupId#2]
+--LogicalJoin(type=INNER_JOIN) [GroupId#10]
|--LogicalProject(...)
| +--LogicalJoin(type=LEFT_SEMI_JOIN)
| ...
+--LogicalOlapScan(...)
```
**AFTER APPLY RULE: MERGE_PROJECTS**
```
LogicalProject3( projects=[c_custkey#0, c_name#1]) [should be in GroupId#1, but in GroupId#2 in fact]
+--LogicalJoin(type=INNER_JOIN) [GroupId#10]
|--LogicalProject(...)
| +--LogicalJoin(type=LEFT_SEMI_JOIN)
| ...
+--LogicalOlapScan(...)
```
Since we have exaclty GroupExpression(LogicalProject3 and LogicalProject2) in GroupId#1 and GroupId#2, we need to do MergeGroup(GroupId#1, GroupId#2). But we have child of GroupId#1 in GroupId#2. So the merge is denied.
If the best GroupExpression in GroupId#2 is LogicalProject3, we will get two consecutive projects in the final plan.
There are several configs related to tcmalloc, users do know how to config them. Actually users just want two modes, performance or compact, in performance mode, users want doris run query and load quickly while in compact mode, users want doris run with less memory usage.
If we want to config tcmalloc individually, we can use env variables which are supported by tcmalloc.
When upgrade 1.2 version from 1.1, FE version will don't match BE version for a period of time. After upgrade BE and doing schema change, BE will use a field desc_tbl that add in 1.2 version FE. BE will coredump because the field desc_tbl is nullptr. So it need to refuse the request.
- this pr is used to fix the be core dump when import array.
- before the change, we import array by rapidjson string will core dump under the non-vectorized scenario.
- after the change, we can import array by rapidjson string successfully.
