CK GEMM Int8 Bug Fixes

[turneram]

Adds workarounds to avoid passing capture ops and scalar literals from quantization as arguments to ck_gemm.

[codecov]

Codecov Report

Merging #2229 (f40992f) into develop (e748657) will decrease coverage by 0.02%.
Report is 3 commits behind head on develop.
The diff coverage is 50.00%.

❗ Current head f40992f differs from pull request most recent head ff051a1. Consider uploading reports for the commit ff051a1 to get more accurate results

@@             Coverage Diff             @@
##           develop    #2229      +/-   ##
===========================================
- Coverage    91.37%   91.34%   -0.02%     
===========================================
  Files          439      439              
  Lines        16493    16499       +6     
===========================================
+ Hits         15069    15071       +2     
- Misses        1424     1428       +4     

Files	Coverage Δ
src/rewrite_quantization.cpp	86.96% <50.00%> (-8.04%)	⬇️

[migraphx-bot]

Test	Batch	Rate new 4bec7e	Rate old e74865	Diff	Compare
torchvision-resnet50	64	2,847.28	2,846.63	0.02%	✅
torchvision-resnet50_fp16	64	6,476.76	6,403.52	1.14%	✅
torchvision-densenet121	32	2,105.79	2,056.65	2.39%	✅
torchvision-densenet121_fp16	32	3,685.76	3,516.87	4.80%	🔆
torchvision-inceptionv3	32	1,594.31	1,596.40	-0.13%	✅
torchvision-inceptionv3_fp16	32	2,590.69	2,588.49	0.09%	✅
cadene-inceptionv4	16	706.91	707.22	-0.04%	✅
cadene-resnext64x4	16	697.71	697.64	0.01%	✅
slim-mobilenet	64	8,353.52	8,347.67	0.07%	✅
slim-nasnetalarge	64	226.78	226.74	0.02%	✅
slim-resnet50v2	64	2,677.38	2,677.44	-0.00%	✅
bert-mrpc-onnx	8	824.04	825.37	-0.16%	✅
bert-mrpc-tf	1	388.84	389.40	-0.15%	✅
pytorch-examples-wlang-gru	1	297.61	299.03	-0.47%	✅
pytorch-examples-wlang-lstm	1	310.09	313.82	-1.19%	✅
torchvision-resnet50_1	1	601.42	600.59	0.14%	✅
torchvision-inceptionv3_1	1	333.16	338.61	-1.61%	✅
cadene-dpn92_1	1	396.01	396.89	-0.22%	✅
cadene-resnext101_1	1	329.65	326.85	0.86%	✅
slim-vgg16_1	1	462.97	463.41	-0.09%	✅
slim-mobilenet_1	1	2,033.05	2,022.62	0.52%	✅
slim-inceptionv4_1	1	217.22	215.35	0.87%	✅
onnx-taau-downsample	1	306.56	306.54	0.00%	✅
dlrm-criteoterabyte	1	21.72	21.72	0.01%	✅
dlrm-criteoterabyte_fp16	1	40.70	40.72	-0.05%	✅
agentmodel	1	5,747.52	5,816.02	-1.18%	✅
unet_fp16	2	55.92	55.97	-0.07%	✅
resnet50v1_fp16	1	930.21	952.91	-2.38%	✅
bert_base_cased_fp16	64	970.13	970.06	0.01%	✅
bert_large_uncased_fp16	32	304.86	304.82	0.02%	✅
bert_large_fp16	1	167.10	167.06	0.02%	✅
distilgpt2_fp16	16	1,277.67	1,278.04	-0.03%	✅

Check results before merge 🔆

[migraphx-bot]

    :white_check_mark:bert-mrpc-onnx: PASSED: MIGraphX meets tolerance

    :white_check_mark:bert-mrpc-tf: PASSED: MIGraphX meets tolerance

    :white_check_mark:pytorch-examples-wlang-gru: PASSED: MIGraphX meets tolerance

    :white_check_mark:pytorch-examples-wlang-lstm: PASSED: MIGraphX meets tolerance

    :white_check_mark:torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

🔴torchvision-inceptionv3_1: FAILED: MIGraphX is not within tolerance - check verbose output

    :white_check_mark:cadene-dpn92_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:cadene-resnext101_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-vgg16_1: PASSED: MIGraphX meets tolerance

    :white_check_mark:slim-mobilenet_1: PASSED: MIGraphX meets tolerance

🔴slim-inceptionv4_1: FAILED: MIGraphX is not within tolerance - check verbose output

    :white_check_mark:dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

    :white_check_mark:agentmodel: PASSED: MIGraphX meets tolerance

    :white_check_mark:unet: PASSED: MIGraphX meets tolerance

    :white_check_mark:resnet50v1: PASSED: MIGraphX meets tolerance

🔴bert_base_cased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

🔴bert_large_uncased_fp16: FAILED: MIGraphX is not within tolerance - check verbose output

    :white_check_mark:bert_large: PASSED: MIGraphX meets tolerance

🔴distilgpt2_fp16: FAILED: MIGraphX is not within tolerance - check verbose output