Finish moving period_helper to cython

[jbrockmendel]

• closes period_helper should be in cython #19465

There is probably some cleanup that can be done now that period logic is self-contained.

No evident perf impact.

asv continuous -E virtualenv -f 1.1 master HEAD -b period
[...]
       before           after         ratio
     [776fed3a]       [7cc0d9e8]
+        1.14±0μs      1.28±0.01μs     1.13  period.PeriodProperties.time_property('M', 'qyear')
-        579±30ms          509±4ms     0.88  groupby.Datelike.time_sum('period_range')

(will post results from a couple more runs as they become available)
update two more runs show zero change, not even noise.

update Oh, also did some unrelated modernizing of string formatting in timedeltas.

[gfyoung]

https://travis-ci.org/pandas-dev/pandas/jobs/412170226#L3465

Looks like you have some linting failures.

[codecov]

Codecov Report

Merging #22196 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master   #22196   +/-   ##
=======================================
  Coverage   92.06%   92.06%           
=======================================
  Files         169      169           
  Lines       50694    50694           
=======================================
  Hits        46671    46671           
  Misses       4023     4023

Flag	Coverage Δ
#multiple	90.47% <ø> (ø)	⬆️
#single	42.32% <ø> (ø)	⬆️

---

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 776fed3...48e273b. Read the comment docs.

[gfyoung]

https://travis-ci.org/pandas-dev/pandas/jobs/412207881#L3475

Looks like you (still) have some linting failures. 🙁

[gfyoung]

https://travis-ci.org/pandas-dev/pandas/jobs/412219771#L3474

Getting there, but not quite. Are you able to check locally via cpplint?

[jreback]

asv continuous -E virtualenv -f 1.1 master HEAD -b period

can you create an issue to add more period benchmarks

[jbrockmendel]

can you create an issue to add more period benchmarks

Done

[jreback]

thanks!