Rewrite away blockwise Subtensor in gradient of Blockwise(Conv1d)

More stable fix for JAX Multinomial

Fix nan in jax implementation of Multinomial (#1328)

Implement batched convolve1d (#1318)

Co-authored-by: Jesse Grabowski <jessegrabowski@gmail.com>
Co-authored-by: Rob Zinkov <zaxtax@users.noreply.github.com>

Enable new `assume_a` in Solve

Do not introduce 0 strides for broadcastable dimensions in DimShuffle

Some poorly implemented BLAS operations don't handle them correctly

Skip failing JAX test

Started failing in 0.4.36: jax-ml/jax#26888

Skip failing JAX test

Started failing in 0.4.36 and seems to be fixed in 0.5.1

Fix bug in `local_reshape_to_dimshuffle`

Fix bug in JAX test

Fix bug in einsum

A shortcut in the numpy implementation of einsum_path when there's nothing to optimize, creates a default path that can combine more than 2 operands. Our implementation only works with 2 or 1 operand operations at each step.

https://github.com/numpy/numpy/blob/cc5851e654bfd82a23f2758be4bd224be84fc1c3/numpy/_core/einsumfunc.py#L945-L951