loss_fun parameter validationf for entropic GW functions

[kachayev]

Types of changes

• explicitly check if loss_fun given is valid for all GW functions that take such the argument
• fix docstrings where it was still listed as a "callable" (not an option now, based on the code)
• update tests to check for validation

Motivation and context / Related issue

Unified approach to handling loss_fun parameter for all GW-related functions. Cleaner error message.

Also, note that before the changes, oftentimes function relied on the validation to happen rather deep in the call stack, for example in init_matrix helper. Which seems rather fragile when making changes in high-level modules (e.g. another set of low-level procedures could be used).

How has this been tested (if it applies)

Test cases are updated.

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.

[codecov]

Codecov Report

Merging #515 (2a74966) into master (7fa8438) will decrease coverage by 0.02%.
The diff coverage is 88.88%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #515      +/-   ##
==========================================
- Coverage   96.06%   96.05%   -0.02%     
==========================================
  Files          65       65              
  Lines       13706    13732      +26     
==========================================
+ Hits        13167    13190      +23     
- Misses        539      542       +3