The header question seems different than your body question. I am assuming that you are asking how to only get a single CSV file when writing?

To do so you should use the coalesce:

df.coalesce(1).write.format("csv").mode("overwrite").save(dstPath)

This will save a single CSV file underneath the directory that you provide. If you want to have a specific name of the file then you will need to rename it. You could just use dbutils.fs.cp() to copy the file with a new name or you can use Python "os" library to rename it.