Sqoop only runs map only job and there is no reducer initialized in SQOOP.
I don't think there is a way to run reducer phase in Sqoop job.
if you are thinking to get only two files into HDFS then use
--num-mappers 2 //sqoop job with 2 mapper phases
Another possible way to initialize reducer task would be:
If you are thinking to run two reducer phases then import the data into staging directory then trigger an hive job by setting two reducer using
set hive.exec.reducers.bytes.per.reducer=1000000; //change the number accordingly to your requirements.
by using distribute by sort by in the insert query
to insert data from staging table to final table.