Support for replacing date partitioned tables

[paoloburelli]

I addressed the enhancement request in #43, now it is possible to replace data in a partition.
A few other minor changes have been added to give a full support of date partitioned tables (creation and insertion were missing mainly).
I also added a copy function, which is a convenient (and free) way to move data in BigQuery.

[codecov-io]

Codecov Report

Merging #85 into master will decrease coverage by 47.23%.
The diff coverage is 10.32%.

@@             Coverage Diff             @@
##           master      #85       +/-   ##
===========================================
- Coverage   73.61%   26.37%   -47.24%     
===========================================
  Files           4        4               
  Lines        1554     1725      +171     
===========================================
- Hits         1144      455      -689     
- Misses        410     1270      +860

Impacted Files	Coverage Δ
pandas_gbq/tests/test_gbq.py	26.48% <14.77%> (-56.51%)	⬇️
pandas_gbq/gbq.py	17.44% <6.25%> (-58.12%)	⬇️

---

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 79c9067...ded898b. Read the comment docs.

[parthea]

@paoloburelli Firstly, thanks for this PR! I think the next step for pandas-gbq is to leverage the work done in google-cloud-bigquery. I'd like to focus on migrating to google-cloud-bigquery before reviewing this PR since it already has support for some of the features proposed including date partitioned tables and table copy operations. @jasonqng has done a great job getting this task started in PR #25. @GregCT has also expressed interest in continuing the work done in PR #25 and was recently added collaborator (based on the last comment in the PR).

I really appreciate the time that you put into this PR and I want to revisit it after #25 is merged because I agree these features would be helpful for the pandas-gbq community. If you're interested in helping out with #25, please let @jasonqng & @GregCT by commenting in PR #25.

cc @tswast
cc @jreback

[paoloburelli]

Hi, my team currently needs these changes, so we will continue to develop the library on our fork.
Meanwhile I will close this pull request and I'll monitor the development of the library.

[hayato1986]

Hope this goes to next step!

[BartVanDerVurst]

Will this be merged to the master soon? Still an open topic which results in errors when running data-partioning append's

[tswast]

@BartVanDerVurst there have been many changes to the pandas-gbq library since this PR was first created. For one, it now uses the google-cloud-bigquery library under the covers. It would need to be rebased on a more recent version of the library for us to merge it into master.