fix(test): fix failing latitude test

[matthewmayer]

fix #2080

[codecov]

Codecov Report

Merging #2116 (d41c2a6) into next (ec1ca12) will increase coverage by 0.00%.
The diff coverage is 100.00%.

❗ Current head d41c2a6 differs from pull request most recent head 875ef47. Consider uploading reports for the commit 875ef47 to get more accurate results

Additional details and impacted files

@@           Coverage Diff           @@
##             next    #2116   +/-   ##
=======================================
  Coverage   99.59%   99.60%           
=======================================
  Files        2567     2567           
  Lines      243394   243394           
  Branches     1254     1255    +1     
=======================================
+ Hits       242420   242422    +2     
+ Misses        947      945    -2     
  Partials       27       27           

Impacted Files	Coverage Δ
src/locales/en_AU_ocker/location/city_name.ts	100.00% <ø> (ø)
src/locales/en_NG/location/city_name.ts	100.00% <ø> (ø)
src/locales/en_ZA/location/city_name.ts	100.00% <ø> (ø)
src/locales/en_AU_ocker/location/city_pattern.ts	100.00% <100.00%> (ø)
src/locales/en_AU_ocker/location/index.ts	100.00% <100.00%> (ø)
src/locales/en_NG/location/city_pattern.ts	100.00% <100.00%> (ø)
src/locales/en_NG/location/index.ts	100.00% <100.00%> (ø)
src/locales/en_ZA/location/city_pattern.ts	100.00% <100.00%> (ø)
src/locales/en_ZA/location/index.ts	100.00% <100.00%> (ø)

... and 2 files with indirect coverage changes

[xDivisionByZerox]

This doesn't really fix the issue, since those test failures come from number.float returning a floating point number where all decimal places are exactly 0.
For example faker.number.float({ precision: 0.001 }) CAN return a number like x.000 which JS trims to x. This is not a problem with the float implementation, but with how JS handles floating point number.

The fix you provided will have nicer error logs (at least I believe so) but doesn't actually fix the problem.

Ok, I read your comments in #2080 and saw that you were already aware of the actual problem. That made me curious and have an additional look at the chang you provided. I didn't see that the precision function was actually used to change the expected value. MB.