The following changes are included:
- Starting from TF2.13v, TensorFlow uses Clang as compiler for Linux. Thus, we update the toolchain in release_cpu_linux and release_gpu_linux configs.
- Preserve the old linux build options in the unsupported_cpu_linux and unsupported_gpu_linux configs. If your project fails to build with Clang, you can use these unsupported flags to replace the release flags in your build command. However, please note that the old toolchain is no longer officially supported by TensorFlow and the unsupported configs will be removed soon. We strongly recommend that you migrate to Clang as your compiler for TensorFlow Linux builds. Instructions are available in the official documentation: https://www.tensorflow.org/install/source#install_clang_recommended_linux_only. Another good alternative is to use our Docker containers to build and test TensorFlow: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/tools/tf_sig_build_dockerfiles.
- Add official linker options and container environment settings to release linux configs.
- Deduplicate build options in cpu.bazelrc and gpu.bazelrc.
- Delete outdated CI jobs.

PiperOrigin-RevId: 564620926

PiperOrigin-RevId: 564589017

Imported from GitHub PR https://github.com/openxla/xla/pull/5498

Remove unnecessary padding for vector bias for non-fp8 cublasLt Matmul rewrite.
Copybara import of the project:

--
fb25577449faf13ef424a827b4896a6f7e2c0f53 by wenscarl <shuw@nvidia.com>:

Add CollectivePermute to recursive pattern matching whitelist.

--
b02af4b60279933d4a19ab54b9f4e9291d418210 by shuw <shuw@nvidia.com>:

Fix vector bias padding for f32

Merging this change closes #5498

PiperOrigin-RevId: 564585859

#tf-data The default value of `warm_start` should be `None` instead of `False`. Setting it to `False` as the default value may break the `tf.data.Options` propagation when users set it to `True` somewhere in the input pipeline.

PiperOrigin-RevId: 564585601

#tf-data Output of a model node can be deleted by a thread while some of its input nodes are in the process of being deleted. If such a case happens, the `output_` of these input nodes becomes a dangling pointer. This CL adds a check for such a scenario.

PiperOrigin-RevId: 564583894

PiperOrigin-RevId: 564578779

PiperOrigin-RevId: 564577405

These patterns are used only by TFLite which now has direct support for stablehlo.scatter.

PiperOrigin-RevId: 564569917

PiperOrigin-RevId: 564559805

Sometimes scales are zero in the all zero case. This is obviously
silly but it triggers a failure in the pipeline. Make scale the
minimum scale value as whatevs.

PiperOrigin-RevId: 564556649

PiperOrigin-RevId: 564551916

Add the custom call xla.gpu.init_cuBLAS to the beginning of the entry function of the module, when the module contains a gemm.

PiperOrigin-RevId: 564550744

The user will provide inputs via AHWB. Behind the scenes we bind the AHWB's to SSBO. Both cl & gl inference are supported.

PiperOrigin-RevId: 564547199

PiperOrigin-RevId: 564546615

PiperOrigin-RevId: 564546472

We are not going to pursue direct lowering from LMHLO to API calls, though we might consider "reference implementation API" in the future, but it will go through intermediate "XLA runtime dialect". For now remove dead code.

PiperOrigin-RevId: 564538039

Leaky Relu, Elu, Relu, Relu1, Relu0To1, Relu6, Tanh, Sigmoid, and HardSwish

PiperOrigin-RevId: 564537536

PiperOrigin-RevId: 564529533

PiperOrigin-RevId: 564528951

Needed for sharding int4 models.

PiperOrigin-RevId: 564528408

PiperOrigin-RevId: 564526562

PiperOrigin-RevId: 564526383

PiperOrigin-RevId: 564518057

PiperOrigin-RevId: 564516421

Rearrange the code that unroll the last while-loop iteration to avoid cloning
the instruction being pipelined, which is unused in the iteration. This is to
prepare for the next change to support backward pipelining recv-done without
generating the unused recv and recv-done pair.

PiperOrigin-RevId: 564515742

1. Adding bidirectional propagation paths among nodes within the same shard_group
2. Enforce the shardings from the same shard_as group to take the same sharding in the end.

PiperOrigin-RevId: 564504486

PiperOrigin-RevId: 564500466

PiperOrigin-RevId: 564494782

According to
https://www.tensorflow.org/api_docs/python/tf/math/unsorted_segment_sum:

> If the given segment ID i is negative, the value is dropped and will not be
> added to the sum of the segment.

PiperOrigin-RevId: 564485570

PiperOrigin-RevId: 564472190

Leaky Relu, Elu, Relu, Relu1, Relu0To1, Relu6, Tanh, Sigmoid, and HardSwish

PiperOrigin-RevId: 564472058

[SPMD] Create a placeholder sharding of type UNKNOWN with the lowest precedence that can be overwriten by any other shardings. (2/N)

UNKNOWN type is not supported in subgroup sharding.

PiperOrigin-RevId: 564456004

This way libtpu can use symbols from it.

PiperOrigin-RevId: 564454296

PiperOrigin-RevId: 564441419

Also
- Unify naming to ${scope}_${thing}_dim_idx.
- Make logic for split_lhs_nc a bit easier to follow (at least in my opinion), by
  decoupling it from the batch_size computation.

PiperOrigin-RevId: 564435916

PiperOrigin-RevId: 564434713

PiperOrigin-RevId: 564428072

[XLA] Use max size of buffer in heap simulator and buffer assignment and make size DCHECKs into CHECKs

PiperOrigin-RevId: 564421004

PiperOrigin-RevId: 564420754

PiperOrigin-RevId: 564415721