Extend weight compression with INT8 symmetric scheme

[l-bat]

Changes

Added INT8_SYM compression mode

Reason for changes

INT8_SYM mode can provide a better performance and required for dynamic quantization

Related tickets

124823

Tests

Updated tests/openvino/native/quantization/test_weights_compression.py

[codecov]

Codecov Report

Merging #2288 (479d4f4) into develop (08cd0f7) will decrease coverage by 4.80%.
Report is 8 commits behind head on develop.
The diff coverage is 100.00%.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #2288      +/-   ##
===========================================
- Coverage    90.76%   85.96%   -4.80%     
===========================================
  Files          486      487       +1     
  Lines        43764    43805      +41     
===========================================
- Hits         39723    37659    -2064     
- Misses        4041     6146    +2105     

Flag	Coverage Δ
COMMON	15.86% <10.81%> (+0.05%)	⬆️
ONNX	?
OPENVINO	38.71% <89.18%> (+0.02%)	⬆️
TENSORFLOW	30.01% <10.81%> (-0.01%)	⬇️
TORCH	62.70% <24.32%> (+0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
nncf/parameters.py	100.00% <100.00%> (ø)
...ization/algorithms/weight_compression/algorithm.py	95.91% <ø> (ø)
...ntization/algorithms/weight_compression/backend.py	100.00% <ø> (ø)
.../algorithms/weight_compression/openvino_backend.py	98.23% <100.00%> (-0.03%)	⬇️
nncf/quantization/quantize_model.py	82.89% <100.00%> (-1.83%)	⬇️
nncf/torch/quantization/quantize_model.py	91.42% <100.00%> (ø)

... and 44 files with indirect coverage changes

[ljaljushkin]

Special thanks for refactoring and fixing issue with printing statistics

[alexsu52]

What do you think about leaving INT8 as an alias for INT8_ASYM?