Use pandas duplicated() to check for duplicate index values

[victorlin]

Description of proposed changes

This PR uses the built-in pandas.Series.duplicated() to check for duplicates, based on @huddlej's original suggestion. Also cleans up the logic for optimal performance.

Related issue(s)

Fixes #825

Testing

• Added test

[codecov]

Codecov Report

Merging #826 (73ccb0e) into master (73a07cf) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #826      +/-   ##
==========================================
+ Coverage   33.78%   33.79%   +0.01%     
==========================================
  Files          41       41              
  Lines        5902     5903       +1     
  Branches     1465     1465              
==========================================
+ Hits         1994     1995       +1     
  Misses       3825     3825              
  Partials       83       83              

Impacted Files	Coverage Δ
augur/util_support/metadata_file.py	100.00% <100.00%> (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 73a07cf...73ccb0e. Read the comment docs.

[victorlin]

[side note] While working on this, I opened a performance bug in pandas pandas-dev/pandas#45236 related to this line:

augur/augur/util_support/metadata_file.py

Line 59 in 73ccb0e

duplicate_rows = self.metadata[self.key_type].duplicated()

The docs suggest using self.metadata.duplicated(self.key_type), but it is actually slower.

[huddlej]

Thanks, @victorlin! I confirmed that this works with the ncov scripts that call read_metadata when I have numexpr installed.