feat: support read and write from hive datasource

[awang12345]

What type of PR is this?

• feature

What problem(s) does this PR solve?

Issue(s) number:

Description:

Add hive datasource to read and write

How do you solve it?

hive: {
# algo's data source from hive
read: {
#[Optional] spark and hive require configuration on different clusters
metaStoreUris: "thrift://hive-metastore-server-01:9083"
#spark sql
sql: "select column_1,column_2,column_3 from database_01.table_01 "
#[Optional] graph source vid mapping with column of sql result.
srcId: "column_1"
#[Optional] graph dest vid mapping with column of sql result
dstId: "column_2"
#[Optional] graph weight mapping with column of sql result
weight: "column_3"
}

  # algo result sink into hive
  write: {
    #[Optional] spark and hive require configuration on different clusters
    metaStoreUris: "thrift://hive-metastore-server-02:9083"
    #save result to hive table
    dbTableName: "database_02.table_02"
    #[Optional] spark dataframe save mode，optional of Append,Overwrite,ErrorIfExists,Ignore. Default is Overwrite
    saveMode: "Overwrite"
    #[Optional] if auto create hive table. Default is true
    autoCreateTable: true
    #[Optional] algorithm result mapping with hive table column name. Default same with column name of algo result dataframe
    resultTableColumnMapping: {
      # Note: Different algorithms have different output fields, so let's take the pagerank algorithm for example:
      _id: "column_1"
      pagerank: "pagerank_value"
    }
  }
}

Special notes for your reviewer, ex. impact of this fix, design document, etc:

Spark and hive have no environment validation on different clusters. All other cases have been verified

[Nicole00]

please update the comment to English~

[Nicole00]

no need to show.

[Nicole00]

ditto

[awang12345]

done

[Nicole00]

LGTM