Kron typo fixed

[ebelnikola]

Hi!

I encountered an issue with automatic differentiation. When I tried to differentiate a piece of code that involved taking multiple traces of a single tensor, I got an error. Here is a minimal failing example:

using TensorOperations
using Zygote

A = rand(2, 2, 2, 2)

function tst(x)
    @tensoropt res = x[1, 1, 2, 2]
    return sum(res)
end

withgradient(tst, A)

The issue was in the _kron function, which had a typo in the argument order for the call to tensorproduct.

This PR fixes the issue and adds a test to "ad.jl" to verify that the tensortrace! rrule works correctly when taking more than one trace.

[github-actions]

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

[lkdvos]

Thank you so much for the report as well as the fix, this is enormously appreciated!
I took the liberty of quickly running the formatter (we've switched to using Runic.jl these days), but otherwise this looks great and will merge as soon as tests pass.

[ebelnikola]

Thanks!

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/utils.jl	100.00% <100.00%> (+100.00%)	⬆️

... and 30 files with indirect coverage changes

🚀 New features to boost your workflow:

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