[CPU] Fixed 2nd token latency for FC w/o dynamic quantization

[dmitry-gorokhov]

Details:

• This PR fixes 2nd token latency regression on FC w/o dynamic quantizations caused by [CPU] Optimize DQ FC with stack allocated aux accum registers #28576,
• The issue was connected with too agressive unrolling trategy which resulted in increased L1i cache misses.
• oneDNN PR: [FORK][FIX] Limit microkernel unrolling factor oneDNN#275

Tickets:

• CVS-162857

[github-actions]

This PR will be closed in a week because of 2 weeks of no activity.

[github-actions]

This PR was closed because it has been stalled for 2 week with no activity.