This is a patch release containing the following changes to v3.6.1:

• Fixed segmentation fault issue in convolution primitive on processors with Intel AVX2 instruction set support (2eb3dd1)
• Added a workaround for build issue with GCC 8.2 and GNU binutils 2.27 (19ef223, 262fb02, e3782e8)
• Fixed a thread safety issue in matmul primitive for builds relying on Arm Compute Library (ACL) and bumped minimal supported ACL version to 24.11.1 (4d962e7)
• Suppressed spurious warnings for GCC (7d3164d, c805a50, e526172, dc780cb)
• Fixed segfaults in BRGEMM-based matmul, convolution, and deconvolution implementations on AArch64-based processors (a873a1c, 9a1dc92)
• Fixed performance regression in bf16 convolution with ACL on AArch64-based processors (4793296)
• Fixed an issue with convolution primitive creation with PREFER_YMM CPU ISA hint on AArch64-based processors (e34d992)
• Improved bf16 matmul performance with fp32 destination with ACL on AArch64-based processors (548d5d6)
• Improved bf16 to fp32 reorder performance on AArch64-based Processors (917dd13)
• Fixed issue in matmul primitive with 4D tensors on AArch64-based processors (d13c966)
• Suppressed spurious GCC warnings in deconvolution primitive on AArch64-based processors (f90f60e)
• Fixed warnings in BRGEMM implementation on AArch64-based processors (866b196)
• Fixed correctness issue in reorder primitive with zero points for 4D shapes on AArch64-based Processors (836ea10)
• Improved bf16 reorder performance on AArch64-based Processors (12bafbe)
• Fixed performance regression for backward convolution primitive descriptor creation time on Intel processors (2b3389f)
• Improved performance of fp16 matmul with int4 weights on Intel GPUs based on Xe2 architecture (4c8fb2c, 3dd4f43, 280bd28)
• Fixed performance regression for int8 convolution with large spatial sizes on processors with Intel AMX support (05d68df)
• Restricted check for microkernel fusion support to cases when fusion functionality is actually used on Intel GPUs (48f6bd9)