Fix DoRA extraction crash on 1-D weights (x_pad_token, norm weights)

[Booyaka101]

Fixes #15.

Extracting a DoRA from models that contain 1-D weights (e.g. z-image's x_pad_token, or LayerNorm/RMSNorm .weight) crashes with:

Dimension out of range (expected to be in range of [-1, 0], but got 1)
  nodes/dora_extract_wd.py -> torch.linalg.norm(tensor_a, dim=1, keepdim=True)

In the matched-shape branch, is_conv_dim = tensor_a.ndim == 4 sends every non-4-D tensor to torch.linalg.norm(..., dim=1). A 1-D tensor has no dim=1, so it raises and aborts the whole extraction. The function already discards 1-D tensors a few lines later (the weight_diff.ndim < 2 skip), so this just bails out before the norm instead.

Verified with a CPU repro of the exact branch:

1-D input (x_pad_token / norm.weight):  before -> IndexError (dim 1) ;  after -> skipped
2-D input:                              still processed
4-D conv input:                         still processed