Fuse transposes in pointwise and reduce fusions

[pfultz2]

This uses the shape_transform_descriptor to track how the axes are changed throughout the transformation. This adds method to get the common dimensions and generate operators for different points of the rewrite.

[codecov]

Codecov Report

Attention: Patch coverage is 93.26425% with 13 lines in your changes missing coverage. Please review.

Project coverage is 92.17%. Comparing base (f9e276b) to head (e1a768f).
Report is 4 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/shape_transform_descriptor.cpp	92.02%	13 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3705      +/-   ##
===========================================
- Coverage    92.18%   92.17%   -0.02%     
===========================================
  Files          514      514              
  Lines        21780    21891     +111     
===========================================
+ Hits         20078    20177      +99     
- Misses        1702     1714      +12     

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

[migraphx-bot]

Test	Batch	Rate new a502b6	Rate old 6f6705	Diff	Compare
torchvision-resnet50	64	3,256.53	3,257.99	-0.04%	✅
torchvision-resnet50_fp16	64	7,002.67	6,969.89	0.47%	✅
torchvision-densenet121	32	2,433.48	2,434.72	-0.05%	✅
torchvision-densenet121_fp16	32	4,065.21	4,079.58	-0.35%	✅
torchvision-inceptionv3	32	1,627.30	1,625.71	0.10%	✅
torchvision-inceptionv3_fp16	32	2,746.69	2,744.57	0.08%	✅
cadene-inceptionv4	16	764.29	764.53	-0.03%	✅
cadene-resnext64x4	16	812.75	812.65	0.01%	✅
slim-mobilenet	64	7,468.00	7,463.64	0.06%	✅
slim-nasnetalarge	64	209.05	209.02	0.02%	✅
slim-resnet50v2	64	3,445.62	3,438.27	0.21%	✅
bert-mrpc-onnx	8	1,146.16	1,150.14	-0.35%	✅
bert-mrpc-tf	1	473.35	473.42	-0.01%	✅
pytorch-examples-wlang-gru	1	415.52	412.17	0.81%	✅
pytorch-examples-wlang-lstm	1	391.81	393.10	-0.33%	✅
torchvision-resnet50_1	1	775.66	790.72	-1.91%	✅
cadene-dpn92_1	1	396.75	399.49	-0.69%	✅
cadene-resnext101_1	1	382.25	383.53	-0.33%	✅
onnx-taau-downsample	1	345.29	345.20	0.03%	✅
dlrm-criteoterabyte	1	33.32	33.32	-0.01%	✅
dlrm-criteoterabyte_fp16	1	52.72	52.75	-0.05%	✅
agentmodel	1	8,175.55	8,373.46	-2.36%	✅
unet_fp16	2	58.37	58.79	-0.72%	✅
resnet50v1_fp16	1	925.29	969.89	-4.60%	🔴
resnet50v1_int8	1	1,016.74	1,031.16	-1.40%	✅
bert_base_cased_fp16	64	1,168.79	1,170.20	-0.12%	✅
bert_large_uncased_fp16	32	363.13	362.80	0.09%	✅
bert_large_fp16	1	199.97	199.81	0.08%	✅
distilgpt2_fp16	16	2,200.15	2,195.15	0.23%	✅
yolov5s	1	532.13	534.95	-0.53%	✅
tinyllama	1	43.37	43.65	-0.65%	✅
vicuna-fastchat	1	173.99	174.76	-0.44%	✅
whisper-tiny-encoder	1	417.20	417.83	-0.15%	✅
whisper-tiny-decoder	1	428.18	425.41	0.65%	✅

This build is not recommended to 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