The DataSet merge method exist bug

[Linary]

Please check the FAQ documentation before raising an issue

Describe the bug (required)

The merge method of DataSet exist bug:

• colNames insert at the end position of origin colNames;
• rowValues insert at the begin position of origin rowValues;

  // merge two DataSet Horizontally with same row count
  bool merge(DataSet&& o) {
    if (rowSize() != o.rowSize()) {
      return false;
    }
    auto newColSize = colSize() + o.colSize();
    colNames.reserve(newColSize);
    // at end
    colNames.insert(colNames.end(),
                    std::make_move_iterator(o.colNames.begin()),
                    std::make_move_iterator(o.colNames.end()));
    for (std::size_t i = 0; i < rowSize(); ++i) {
      rows[i].values.reserve(newColSize);
      // at begin
      rows[i].values.insert(rows[i].values.begin(),
                            std::make_move_iterator(o.rows[i].values.begin()),
                            std::make_move_iterator(o.rows[i].values.end()));
    }
    return true;
  }

Your Environments (required)

• OS: uname -a
• Compiler: g++ --version or clang++ --version
• CPU: lscpu
• Commit id (e.g. a3ffc7d8)

How To Reproduce(required)

Steps to reproduce the behavior:

1. Step 1 : define DataSet1

nebula::DataSet data1({"col1", "col2", "col3"});
data1.emplace_back(nebula::Row({1, 2, 3}));

2. Step 2 : define DataSet2

nebula::DataSet data2({"col4", "col5"});
data2.emplace_back(nebula::Row({4, 5}));

3. Step 3 : merge DataSet1 and DataSet2

data1.merge(std::move(data2));

Expected behavior

data1 should be

col1	col2	col3	col4	col5
1	2	3	4	5

Actual behavior

data1 is

col1	col2	col3	col4	col5
4	5	1	2	3

Additional context