Multilingual regression fixes

[lintool]

@Impavidity this is all on tuna.

This one works fine:

nohup python src/main/python/run_regression.py --collection clef06-fr --index >& log.clef06-fr &

These work fine, but I caught a few typos:

nohup python src/main/python/run_regression.py --collection fire12-bn --index >& log.fire12-bn &
nohup python src/main/python/run_regression.py --collection fire12-hi --index >& log.fire12-hi &
nohup python src/main/python/run_regression.py --collection fire12-en --index >& log.fire12-en &

This appears to be broken:

nohup python src/main/python/run_regression.py --collection trec02-ar --index >& log.trec02-ar &

Also, the generated regression docs are different.

Can you please check?

[codecov]

Codecov Report

Merging #902 into master will increase coverage by 0.01%.
The diff coverage is n/a.

@@             Coverage Diff              @@
##             master     #902      +/-   ##
============================================
+ Coverage     36.01%   36.02%   +0.01%     
- Complexity      474      475       +1     
============================================
  Files           119      119              
  Lines          7139     7139              
  Branches       1082     1082              
============================================
+ Hits           2571     2572       +1     
  Misses         4289     4289              
+ Partials        279      278       -1

Impacted Files	Coverage Δ	Complexity Δ
...anserini/ltr/feature/base/PMIFeatureExtractor.java	86.53% <0%> (+1.92%)	13% <0%> (+1%)	⬆️

---

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 dd47b91...ccf339e. Read the comment docs.

[lintool]

@Impavidity verified that AR fixed.

[Impavidity]

LGTM.