Working contract_network

[JoeyT1994]

This PR introduces the ability to contract vectors of AbstractArrays and thus AbstractTensorNetworks via the contractnetwork(tn; sequence) function.

The sequence is used to tell the contraction_sequence backend how to find the desired sequence, with the strings sequence= "leftassociative" and "optimal" supported (the latter via TensorOperations) alongside support for sending a specific pre-defined sequence.

Rudimentary tests are included, although we might want to consider making these more robust and also testing the sequence finding functionality ourselves.

[codecov]

Codecov Report

❌ Patch coverage is 93.10345% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 56.51%. Comparing base (2819116) to head (5a7dc37).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/contract_network.jl	91.66%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main       #7      +/-   ##
==========================================
+ Coverage   54.13%   56.51%   +2.37%     
==========================================
  Files           4        6       +2     
  Lines         447      476      +29     
==========================================
+ Hits          242      269      +27     
- Misses        205      207       +2     

Flag	Coverage Δ
docs	0.00% <0.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

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

[github-actions]

Your PR no longer requires formatting changes. Thank you for your contribution!

[JoeyT1994]

@mtfishman I'm not really sure what the failing test is to do with... something coming from Aqua.jl and stale dependencies...

[mtfishman]

@mtfishman I'm not really sure what the failing test is to do with... something coming from Aqua.jl and stale dependencies...

It looks like moving the TensorOperations code to a package extension made it so that ITensorBase is no longer a direct dependency of ITensorNetworksNext, since now it is only used in that package extension (Aqua is rightly detecting that situation and seeing that there is a dependency listed in the Project.toml that isn't being used in the main package).

However, in this case I think we should try to use NamedDimsArrays directly rather than ITensorBase, I can give you pointers on how to do that.

[mtfishman]

Could you also bump the package version?

[mtfishman]

Also thinking about it more, I'd slightly prefer the name contract_network over contractnetwork, I think it is easier to read and makes it clearer it is a version of contract for networks.

[JoeyT1994]

Makes sense. I'll make that change.

[mtfishman]

I think this looks good for now, thanks for sticking with all the comments. I think there are a few follow-ups to this:

1. The default keyword argument wrangling is a little bit hard to follow, maybe thinking about it more broadly using Add default_kwargs interface and associated macros #13 could help.
2. As discussed, we should implement an eager contraction sequence finder and use that as the default instead of "leftassociative", I can do that in a follow-up PR.
3. Adding more sequence finding backends to get feature-completeness with ITensorNetworks.jl.

But this seems like a good enough start to build other algorithms like BP so I'll merge this and we can work on those things as next steps.