Fixes to parse DynamicQuantizeLinear

[TedThemistokleous]

• Remove preemptive fold at beginning of parse for inputs
• Fix error with sub/division used for zero point
• add proper reduce_max/min to get scales

Observed 20% increase in runs when using these likely due to removal of flatten.

Separated out from #2826 as we're not observing mismatches off 6.1 and develop anymore for uint8/int8 when dynamicquantizelinear is used

[TedThemistokleous]

@pfultz2 not seeing much of a perf boost with this relative to develop if at all. Bigger perf boost is with the other changeset when we add in the pass.

[migraphx-bot]

Test	Batch	Rate new 6dc8ef	Rate old 21b71c	Diff	Compare
torchvision-resnet50	64	2,858.98	2,861.58	-0.09%	✅
torchvision-resnet50_fp16	64	6,593.89	6,598.71	-0.07%	✅
torchvision-densenet121	32	2,093.84	2,086.95	0.33%	✅
torchvision-densenet121_fp16	32	3,698.50	3,699.24	-0.02%	✅
torchvision-inceptionv3	32	1,600.71	1,604.25	-0.22%	✅
torchvision-inceptionv3_fp16	32	2,573.69	2,573.09	0.02%	✅
cadene-inceptionv4	16	726.86	727.24	-0.05%	✅
cadene-resnext64x4	16	682.69	682.78	-0.01%	✅
slim-mobilenet	64	5,934.99	5,935.93	-0.02%	✅
slim-nasnetalarge	64	153.89	153.98	-0.06%	✅
slim-resnet50v2	64	2,662.44	2,661.75	0.03%	✅
bert-mrpc-onnx	8	918.26	917.95	0.03%	✅
bert-mrpc-tf	1	429.45	423.40	1.43%	✅
pytorch-examples-wlang-gru	1	418.46	424.32	-1.38%	✅
pytorch-examples-wlang-lstm	1	385.86	388.88	-0.78%	✅
torchvision-resnet50_1	1	591.21	601.98	-1.79%	✅
cadene-dpn92_1	1	391.80	393.38	-0.40%	✅
cadene-resnext101_1	1	331.26	331.66	-0.12%	✅
onnx-taau-downsample	1	305.36	304.26	0.36%	✅
dlrm-criteoterabyte	1	28.78	28.76	0.06%	✅
dlrm-criteoterabyte_fp16	1	49.59	49.62	-0.07%	✅
agentmodel	1	7,637.84	7,583.11	0.72%	✅
unet_fp16	2	57.57	57.49	0.13%	✅
resnet50v1_fp16	1	893.90	897.05	-0.35%	✅
resnet50v1_int8	1	822.35	822.69	-0.04%	✅
bert_base_cased_fp16	64	1,055.26	1,056.04	-0.07%	✅
bert_large_uncased_fp16	32	311.87	311.92	-0.02%	✅
bert_large_fp16	1	159.29	159.14	0.09%	✅
distilgpt2_fp16	16	1,859.50	1,857.60	0.10%	✅
yolov5s	1	480.32	476.08	0.89%	✅
tinyllama	1	32.84	32.87	-0.09%	✅
vicuna-fastchat	1	159.72	158.65	0.67%	✅
whisper-tiny-encoder	1	349.13	354.42	-1.49%	✅
whisper-tiny-decoder	1	399.73	399.05	0.17%	✅

This build is OK for merge ✅

[migraphx-bot]

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

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

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

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

     ✅ torchvision-resnet50_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ dlrm-criteoterabyte: PASSED: MIGraphX meets tolerance

     ✅ agentmodel: PASSED: MIGraphX meets tolerance

     ✅ unet: PASSED: MIGraphX meets tolerance

     ✅ resnet50v1: PASSED: MIGraphX meets tolerance

     ✅ bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

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

     ✅ bert_large: PASSED: MIGraphX meets tolerance

     ✅ yolov5s: PASSED: MIGraphX meets tolerance

     ✅ tinyllama: PASSED: MIGraphX meets tolerance

     ✅ vicuna-fastchat: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-encoder: PASSED: MIGraphX meets tolerance

     ✅ whisper-tiny-decoder: PASSED: MIGraphX meets tolerance

     ✅ distilgpt2_fp16: PASSED: MIGraphX meets tolerance

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 91.84%. Comparing base (21b71c6) to head (6dc8ef1).
Report is 3 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop    #2896   +/-   ##
========================================
  Coverage    91.84%   91.84%           
========================================
  Files          478      478           
  Lines        18179    18181    +2     
========================================
+ Hits         16696    16698    +2     
  Misses        1483     1483           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.