jax.nn.selu gradient returns zero for large negative float64 input

Multi-stream windowed einsum fails HLO verifier when a per-device shard > ~1024 — REGRESSION new in jax/jaxlib 0.10.2 (CUDA)

`jit` C++ fastpath attaches the logical instead of the physical sharding to PRNG key outputs, causing a spurious cache miss when keys are fed back in

jax.nn.scaled_matmul raises NotImplementedError on CPU with no graceful error

jax.scipy.special.digamma returns NaN for signed zero and positive subnormal float32 inputs

jax.scipy.stats.chi2.logpdf returns nan at x=0.0 and x=inf for df=2.0

jax.jit produces non-finite gradients for a function involving jnp.cross(x, x) and jnp.linalg.cholesky where eager gradients are finite

jax.scipy.stats.chi2.logpdf returns NaN at x=0 and x=inf when df=2 (inconsistent with gamma.logpdf)

jax.scipy.special.kl_div returns nan instead of inf at x=inf, y=0.0

`jnp.cumsum(x, dtype=bool)` produces wrong values for integer input