Enable mlir input fusion for unary operators and dequantizelinear

[pfultz2]

• Enable mlir input fusion for unary operators and dequantizelinear as there doesnt seem to be a perf regression for these cases. This is also needed for int4.
• Fuse input reshapes as well like we do for the input to gemms.
• Fix a bug where insert_common_op would convert everything to the same type which doesnt work for operators like dequantizelinear. Instead an option is added to disable this behavior.

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 92.85714% with 2 lines in your changes missing coverage. Please review.

Project coverage is 92.17%. Comparing base (bdbe342) to head (a77d3d8). Report is 3 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/common.cpp	92.85%	2 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## develop #3513 +/- ## =========================================== + Coverage 92.08% 92.17% +0.09% =========================================== Files 510 512 +2 Lines 21094 21387 +293 =========================================== + Hits 19424 19714 +290 - Misses 1670 1673 +3 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[migraphx-bot]

Test	Batch	Rate new d0f3e7	Rate old a1e339	Diff	Compare
torchvision-resnet50	64	3,261.21	3,261.15	0.00%	:white_check_mark:
torchvision-resnet50_fp16	64	6,997.04	6,998.48	-0.02%	:white_check_mark:
torchvision-densenet121	32	2,437.41	2,437.74	-0.01%	:white_check_mark:
torchvision-densenet121_fp16	32	4,080.37	4,082.54	-0.05%	:white_check_mark:
torchvision-inceptionv3	32	1,637.76	1,637.99	-0.01%	:white_check_mark:
torchvision-inceptionv3_fp16	32	2,759.67	2,759.69	-0.00%	:white_check_mark:
cadene-inceptionv4	16	776.49	777.37	-0.11%	:white_check_mark:
cadene-resnext64x4	16	811.14	809.89	0.15%	:white_check_mark:
slim-mobilenet	64	7,538.02	7,538.41	-0.01%	:white_check_mark:
slim-nasnetalarge	64	211.59	211.83	-0.11%	:white_check_mark:
slim-resnet50v2	64	3,507.45	3,505.62	0.05%	:white_check_mark:
bert-mrpc-onnx	8	1,150.29	1,153.97	-0.32%	:white_check_mark:
bert-mrpc-tf	1	472.58	463.62	1.93%	:white_check_mark:
pytorch-examples-wlang-gru	1	423.04	416.54	1.56%	:white_check_mark:
pytorch-examples-wlang-lstm	1	402.31	375.99	7.00%	:high_brightness:
torchvision-resnet50_1	1	779.76	787.35	-0.96%	:white_check_mark:
cadene-dpn92_1	1	398.36	398.81	-0.11%	:white_check_mark:
cadene-resnext101_1	1	382.98	382.80	0.05%	:white_check_mark:
onnx-taau-downsample	1	343.41	342.52	0.26%	:white_check_mark:
dlrm-criteoterabyte	1	33.34	33.35	-0.05%	:white_check_mark:
dlrm-criteoterabyte_fp16	1	52.77	52.52	0.47%	:white_check_mark:
agentmodel	1	8,525.44	8,468.76	0.67%	:white_check_mark:
unet_fp16	2	58.89	58.99	-0.17%	:white_check_mark:
resnet50v1_fp16	1	962.49	916.35	5.04%	:high_brightness:
resnet50v1_int8	1	996.79	972.58	2.49%	:white_check_mark:
bert_base_cased_fp16	64	1,171.78	1,170.84	0.08%	:white_check_mark:
bert_large_uncased_fp16	32	363.71	363.63	0.02%	:white_check_mark:
bert_large_fp16	1	200.60	199.03	0.79%	:white_check_mark:
distilgpt2_fp16	16	2,205.04	2,203.30	0.08%	:white_check_mark:
yolov5s	1	535.73	528.92	1.29%	:white_check_mark:
tinyllama	1	43.44	43.76	-0.73%	:white_check_mark:
vicuna-fastchat	1	177.74	173.75	2.30%	:white_check_mark:
whisper-tiny-encoder	1	417.90	418.39	-0.12%	:white_check_mark:
whisper-tiny-decoder	1	429.42	427.20	0.52%	:white_check_mark:

Check results before merge :high_brightness:

[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

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

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

     :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

     :white_check_mark: bert_base_cased_fp16: PASSED: MIGraphX meets tolerance

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

     :white_check_mark: bert_large: PASSED: MIGraphX meets tolerance

     :white_check_mark: yolov5s: PASSED: MIGraphX meets tolerance

     :white_check_mark: tinyllama: PASSED: MIGraphX meets tolerance

     :white_check_mark: vicuna-fastchat: PASSED: MIGraphX meets tolerance

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

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

     :white_check_mark: distilgpt2_fp16: PASSED: MIGraphX meets tolerance