Add aarch64 contributions

Signed-off-by: Siddhartha Menon <[email protected]>

Author: Sqvid

RELEASE_NOTES.md

@@ -29,6 +29,7 @@
 * Improved `fp16` reorder performance.
 * Improved `int8` matmul performance.
 * Improved `bf16` inner product forward propagation performance with Arm Compute Library (ACL).
+* Improved `bf16` eltwise performance.
 * Improved convolution performance on processors with SVE support with ACL.
 
 # Functionality
@@ -68,6 +69,9 @@
 * Enabled support for ROCm 6 on AMD GPUs.
 * Improved CMake integration for oneDNN installation with Nvidia backend enabled.
 * Reduced memory footprint for matmul primitive when using ACL.
+* Fixed correctness issues in `fp32` reorders when using ACL.
+* Fixed sporadic crash in matmuls when using ACL with threadpool scheduler.
+* Fixed sporadic correctness issue in convolutions with ACL.
 
 [`ONEDNN_ENABLE_PRIMITIVE`]: https://uxlfoundation.github.io/oneDNN/v3.8/dev_guide_build_options.html#onednn-enable-primitive