Can get a DDF by sampling a Spark DDF by size

[nhanitvn]

Description and related tickets, documents

Also:

• The resulted DDF has exact 'size' number of rows
• Sample by fraction now return exact number of rows as specified by fraction
• The old sample by fraction is renamed to indicate that the result is approximated.

Reviewers: @phvu @hai-adatao @huan

Breaking changes & backward compatible issues

How to test

Describe how this PR is tested. In case manual testing is required, describe how to do so.

PR Progress

Make sure all checkboxes below are checked before merged

• [x] Branch is in format prefix/description (see this)
• [ ] Merge check has no conflicts. PR checks passed.
• [ ] Code review is done.

[hai-adatao]

1.4.16

[nhanitvn]

Yes, let's this be in 1.4.16 @phvu

[nhanitvn]

This is ready to be merged, @hai-adatao