- Fixed bugs in handling of server-side apply, create, and update API requests for objects containing duplicate items in keyed lists.
  - A `create` or `update` API request with duplicate items in a keyed list no longer wipes out managedFields. Examples include env var entries with the same name, or port entries with the same containerPort in a pod spec.

accurately and efficiently track set membership across `apply` operations. The format used for `ApplySet` is set out in  [KEP 3659](https://github.com/kubernetes/enhancements/issues/3659) as a low-level specification.  Other tools in the ecosystem can also build on this specification for improved interoperability.  To try the ApplySet-based pruning alpha, set `KUBECTL_APPLYSET=true` and use the flags `--prune --applyset=secret-name` with `kubectl apply`. ([#116205](https://github.com/kubernetes/kubernetes/pull/116205),...

* `dependencies`: to include ✨ top-level dependencies ✨ that apply to the whole application. E.g. to add global authentication.
* `callbacks`: OpenAPI callbacks that apply to all the *path operations*.
* `deprecated`: to mark all the *path operations* as deprecated. 🤷
* `include_in_schema`: to allow excluding all the *path operations* from the OpenAPI schema.
* `responses`: OpenAPI responses that apply to all the *path operations*.

For example:

      Value y_out = bn_train_op.getResult(0);
      Value batch_mean = bn_train_op.getResult(1);
      Value reserve_space_1 = batch_mean;
      Value batch_variance = bn_train_op.getResult(2);

      // Apply Bessel's correction on the variance.
      int total_input_size = bn_train_input_type_tensor.getNumElements();
      int total_scale_size = scale_type_tensor.getNumElements();
      int sample_size =

## Changelog since v1.29.0

## Changes by Kind

### Deprecation

- kubectl: Removed the deprecated flag `prune-whitelist` for apply. Please use the flag `prune-allowlist` instead.
   ([#120246](https://github.com/kubernetes/kubernetes/pull/120246), [@pacoxu](https://github.com/pacoxu))

	} else {
		mem := s.newValue3I(ssa.OpPanicBounds, types.TypeMem, int64(kind), idx, len, s.mem())
		s.endBlock().SetControl(mem)
	}
	s.startBlock(bNext)

	// In Spectre index mode, apply an appropriate mask to avoid speculative out-of-bounds accesses.
	if base.Flag.Cfg.SpectreIndex {
		op := ssa.OpSpectreIndex
		if kind != ssa.BoundsIndex && kind != ssa.BoundsIndexU {
			op = ssa.OpSpectreSliceIndex