[MRG] Use the weights w in ot.dist for the cityblock metric

[itzzdev09]

Types of changes

• Bug fix

Description

ot.dist documents that w is used by the cityblock and minkowski metrics, but the cityblock branch never read it and returned the unweighted distance:

import numpy as np, ot
from scipy.spatial.distance import cdist
x, y, w = np.array([[0., 0.]]), np.array([[1., 1.]]), np.array([1., 2.])
ot.dist(x, y, metric="cityblock", w=w)   # [[2.]]
cdist(x, y, metric="cityblock", w=w)     # [[3.]]

The weighted minkowski branch already handled this. cityblock now applies w the same way, in both the use_tensor=True and loop implementations.

How has this been tested

Added test_dist_weighted_cityblock, which compares against scipy.spatial.distance.cdist for both implementations. It fails on master and passes with this change. test/test_utils.py passes.

Checklist

• My change requires a change to the documentation. (not needed)
• I have added tests to cover my changes.
• I have updated the RELEASES.md file.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.85%. Comparing base (10c74a1) to head (7f52892).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #859      +/-   ##
==========================================
- Coverage   96.86%   96.85%   -0.01%     
==========================================
  Files         128      128              
  Lines       26181    26197      +16     
==========================================
+ Hits        25359    25373      +14     
- Misses        822      824       +2     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.