[MRG] Refactor: improve readibility in the convolutional module

[framunoz]

Types of changes

Refactor the ot.bregman._convolutional.py module to improve the readability of the module. These includes:

• Add a function that returns convol_img function.
• Leave the warning message as a constant in the top of the module.
• Create a function for the verbose report.

Motivation and context / Related issue

There is a lot of duplicate code in this module, which can be encapsulated in functions to make it more modular.

How has this been tested (if it applies)

PR checklist

• I have read the CONTRIBUTING document.
• The documentation is up-to-date with the changes I made (check build artifacts).
• All tests passed, and additional code has been covered with new tests.
• I have added the PR and Issue fix to the RELEASES.md file.

[framunoz]

Why do the _convolutional_barycenter2d_log and _convolutional_barycenter2d_debiased_log functions still throw an error if the jax and tensorflow backends are used? I was checking the functions implemented in these modules, and they are implemented.

[codecov]

Codecov Report

Attention: Patch coverage is 97.67442% with 1 line in your changes missing coverage. Please review.

Project coverage is 97.08%. Comparing base (39cd6ec) to head (dadb470).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #709      +/-   ##
==========================================
+ Coverage   97.06%   97.08%   +0.01%     
==========================================
  Files          98       98              
  Lines       19955    19937      -18     
==========================================
- Hits        19370    19355      -15     
+ Misses        585      582       -3