[REVIEW] Make cudf.api.types imports consistent

[galipremsagar]

We usually don't use relative imports in our code-base, this PR makes the import statements consistent.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (branch-21.08@f71c6fe). Click here to learn what that means.
The diff coverage is n/a.

❗ Current head b34d0a3 differs from pull request most recent head 4f24e65. Consider uploading reports for the commit 4f24e65 to get more accurate results

@@               Coverage Diff               @@
##             branch-21.08    #8571   +/-   ##
===============================================
  Coverage                ?   83.00%           
===============================================
  Files                   ?      109           
  Lines                   ?    18215           
  Branches                ?        0           
===============================================
  Hits                    ?    15119           
  Misses                  ?     3096           
  Partials                ?        0           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f71c6fe...4f24e65. Read the comment docs.

[galipremsagar]

@gpucibot merge

[vyasr]

@galipremsagar Sorry I didn't get around to this quite fast enough :) I'm fine with this change. I think Keith, @shwina and briefly discussed the question of using relative imports in cudf when I made the types PR and we couldn't remember why cudf chose not to use them but Keith and Ashwin were sure there was some reason. I think the changes slipped through the PR and I've just been copy-pasting elsewhere. I'm still not convinced that we should be avoiding relative imports on principle, but for the sake of consistency I'm fine sticking to absolute imports until we figure out a plan either way.