Replace memory addresses with UUIDs to enhance the uniqueness of template_name

[ilkilic]

Aims to address #209 by reducing template redefinition issues through the use of UUIDs instead of memory addresses. While not 100% collision-proof, this change improves uniqueness.

[ilkilic]

@anilbey I cannot add you as a reviewer anymore but I would love to get your approval too if you have time

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.25%. Comparing base (5406434) to head (c270d0f).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #210      +/-   ##
==========================================
+ Coverage   91.24%   91.25%   +0.01%     
==========================================
  Files          95       95              
  Lines        5708     5718      +10     
==========================================
+ Hits         5208     5218      +10     
  Misses        500      500              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[AurelienJaquier]

looks good to me, but I don't really see the connection with the issue. Could you explain it to me, @ilkilic ? Have you tried to reproduce the issue and does this PR fix it?

[ilkilic]

When we construct the template we retrieve the memory address and use it as an id for the template_name, however this wasn't unique, leading to the error "a template cannot be redefined". I think Python was reusing the same memory address, which could happen because python's memory management allows addresses to be recycled when objects are created and destroyed. This likely caused neuron to encounter issues when running in the same context. I have just run the failing notebook again with this branch and it fixed the errors

[anilbey]

Hi @ilkilic thanks a lot for addressing it. I agree UUIDs is the proper solution for this.