Fix calculating the report_date in demand_hourly_pa_ferc714

[rousik]

For some reason report_date could be missing for some number of records (0.7%). This addresses that problem by changing the way we align the reports to year boundaries.

[codecov]

Codecov Report

Attention: 13 lines in your changes are missing coverage. Please review.

Comparison is base (548b958) 88.6% compared to head (bf77d1a) 88.6%.
Report is 62 commits behind head on dev.

Additional details and impacted files

@@          Coverage Diff           @@
##             dev   #2999    +/-   ##
======================================
  Coverage   88.6%   88.6%            
======================================
  Files         91      91            
  Lines      10854   10991   +137     
======================================
+ Hits        9618    9749   +131     
- Misses      1236    1242     +6     

Files	Coverage Δ
src/pudl/metadata/fields.py	100.0% <ø> (ø)
src/pudl/transform/ferc714.py	100.0% <100.0%> (ø)
src/pudl/transform/params/ferc1.py	100.0% <ø> (ø)
src/pudl/output/ferc1.py	88.2% <69.2%> (-0.5%)	⬇️
src/pudl/transform/ferc1.py	96.7% <94.8%> (+<0.1%)	⬆️

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[zaneselvans]

Did you happen to run the full FERC-714 ETL locally for all years with these changes?

It looks like there were NaT values included in the 2020 data which got processed by the integration tests in CI, and it passed, so it seems like it will probably work in general. 🤞🏼

[rousik]

Did you happen to run the full FERC-714 ETL locally for all years with these changes?

It looks like there were NaT values included in the 2020 data which got processed by the integration tests in CI, and it passed, so it seems like it will probably work in general. 🤞🏼

I have ran the affected table locally with dagster with etl_fast.yml. I'm expecting this to be low risk WRT possibility of this breaking on additional data.