Support for scalar optimization

[carrotflakes]

Pull Request Template

Checklist

• Confirmed that run-checks all script has been executed.
• Made sure the book is up to date with changes in this PR.

Related Issues/PRs

Provide links to relevant issues and dependent PRs.

Changes

I encountered an error when attempting to optimize scalars (Param<Tensor<B, 0>>). The issue stemmed from the absence of AdaptorRecordV1::Rank0 . This was likely an oversight, so I added Rank0 support.

Testing

I confirmed that it is possible to train my model that include Param<Tensor<B, 0>>.

[carrotflakes]

The test is failing, but I don't think the change is the cause.
Is there anything I can do to fix the test?
Also, is there something missing in this pull request?

[antimora]

CI is failing due to this error in WGPU backend, which could be a fluke. I will re-run the CI again. Meanwhile I am tagging @nathanielsimard and @louisfd:

 test tests::jit_fusion::var::tests::test_var_mean_bias ... ok
  
  failures:
  
  ---- tests::jit::kernel::reduction::reduction::reduction_sum_dim_should_work_with_multiple_invocations stdout ----
  thread 'tests::jit::kernel::reduction::reduction::reduction_sum_dim_should_work_with_multiple_invocations' panicked at crates/burn-wgpu/src/lib.rs:73:5:
  Tensors are not approx eq:
    => Position 1: 507.8968200683594 != 507.8955383300781 | difference 0.00128173828125 > tolerance 0.0010000000000000002
  
  
  failures:
      tests::jit::kernel::reduction::reduction::reduction_sum_dim_should_work_with_multiple_invocations
  
  test result: FAILED. 1341 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out; finished in 50.05s
  
  error: test failed, to rerun pass `-p burn-wgpu --lib`
  Error: Process completed with exit code 101.

[louisfd]

This test's tolerance was changed from 10e-3 to 10e-2 since the creation of this PR. By merging main into your branch you should be fine.

[carrotflakes]

Thank you! merged.