read empty dataframe when one column result by spark iotdb connector

**环境**
- Iotdb 0.9.3 & 0.11.2
- spark 2.3.3
- spark-iotdb-connector 0.11.2

**问题**
spark读取iotdb数据后，返回一个空的DataFrame对象

**分析**
0.9.3中执行带聚合函数的sql，如`select count(root.ln.wf01.wt01.status) from root`，返回为两列，一列是时间列，一列是值，但是在0.11.2中返回的是一列，但是spark-iotdb-connector中并未处理这种，在生成schame的时候，从2开始遍历，导致只有一列的数据返回空schema，代码如下：
spark-iotdb-connector\src\main\scala\org\apache\iotdb\spark\db\Converter.scala
```
def toSparkSchema(options: IoTDBOptions): StructType = {

Class.forName("org.apache.iotdb.jdbc.IoTDBDriver")
val sqlConn: Connection = DriverManager.getConnection(options.url, options.user, options.password)
val sqlStatement: Statement = sqlConn.createStatement()
val hasResultSet: Boolean = sqlStatement.execute(options.sql)

val fields = new ListBuffer[StructField]()
if (hasResultSet) {
val resultSet: ResultSet = sqlStatement.getResultSet
val resultSetMetaData: ResultSetMetaData = resultSet.getMetaData

val printTimestamp = !resultSet.asInstanceOf[IoTDBJDBCResultSet].isIgnoreTimeStamp
if (printTimestamp) {
fields += StructField(SQLConstant.TIMESTAMP_STR, LongType, nullable = false)
}

val colCount = resultSetMetaData.getColumnCount
for (i <- 2 to colCount) {
.....
}
StructType(fields.toList)
}
else {
StructType(fields)
}
}
```
需要加个startCol变量，来确定是从2开始遍历还是1，如下：
```
var startCol = 1
if (printTimestamp) {
fields += StructField(SQLConstant.TIMESTAMP_STR, LongType, nullable = false)
startCol = 2
}

val colCount = resultSetMetaData.getColumnCount
for (i <- startCol to colCount) {
.....
}
```

Start with spark-iotdb-connector/src/main/scala/org/apache/iotdb/spark/db/Converter.scala and inspect toSparkSchema, then reproduce the reported aggregate query with the listed IoTDB and Spark versions. Verify that a one-column result produces a non-empty Spark schema and DataFrame instead of an empty schema.

Written by the indexing model from the issue text.