Enable simplify qdq to work with FP8 types and fix bug in pass

[umangyadav]

match_find_quantizable_ops assumed dequantizelinear only has one use. it fails when it has multiple uses.

This PR fixes that issue and adds FP8 dtypes as allowed quantized type in simplify_qdq pass.

This set of changes helps running FP8 quantized model along with changes in #2506 .

[migraphx-bot]

Test	Batch	Rate new 95616a	Rate old a09dc5	Diff	Compare
torchvision-resnet50	64	2,834.92	2,833.31	0.06%	✅
torchvision-resnet50_fp16	64	6,502.95	6,501.92	0.02%	✅
torchvision-densenet121	32	2,072.36	2,094.20	-1.04%	✅
torchvision-densenet121_fp16	32	3,667.31	3,665.08	0.06%	✅
torchvision-inceptionv3	32	1,598.30	1,594.40	0.24%	✅
torchvision-inceptionv3_fp16	32	2,559.88	2,557.61	0.09%	✅
cadene-inceptionv4	16	722.37	721.79	0.08%	✅
cadene-resnext64x4	16	692.35	692.75	-0.06%	✅
slim-mobilenet	64	8,322.56	8,332.19	-0.12%	✅
slim-nasnetalarge	64	230.55	230.63	-0.03%	✅
slim-resnet50v2	64	2,665.09	2,664.91	0.01%	✅
bert-mrpc-onnx	8	823.76	823.07	0.08%	✅
bert-mrpc-tf	1	388.65	389.27	-0.16%	✅
pytorch-examples-wlang-gru	1	297.92	299.82	-0.63%	✅
pytorch-examples-wlang-lstm	1	313.30	308.52	1.55%	✅
torchvision-resnet50_1	1	608.00	602.66	0.89%	✅
torchvision-inceptionv3_1	1	342.26	343.14	-0.26%	✅
cadene-dpn92_1	1	402.28	404.59	-0.57%	✅
cadene-resnext101_1	1	328.33	328.32	0.00%	✅
slim-vgg16_1	1	459.12	460.67	-0.34%	✅
slim-mobilenet_1	1	2,112.58	2,099.38	0.63%	✅
slim-inceptionv4_1	1	214.86	213.62	0.58%	✅
onnx-taau-downsample	1	305.53	304.86	0.22%	✅
dlrm-criteoterabyte	1	21.59	21.62	-0.14%	✅
dlrm-criteoterabyte_fp16	1	40.63	40.65	-0.06%	✅
agentmodel	1	5,652.14	5,935.95	-4.78%	🔴
unet_fp16	2	54.79	54.68	0.20%	✅
resnet50v1_fp16	1	943.90	932.98	1.17%	✅
bert_base_cased_fp16	64	903.60	903.11	0.06%	✅
bert_large_uncased_fp16	32	285.61	285.63	-0.00%	✅
bert_large_fp16	1	166.63	166.70	-0.04%	✅
distilgpt2_fp16	16	1,281.23	1,281.25	-0.00%	✅

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

     ✅ torchvision-inceptionv3_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-dpn92_1: PASSED: MIGraphX meets tolerance

     ✅ cadene-resnext101_1: PASSED: MIGraphX meets tolerance

     ✅ slim-vgg16_1: PASSED: MIGraphX meets tolerance

     ✅ slim-mobilenet_1: PASSED: MIGraphX meets tolerance

     ✅ slim-inceptionv4_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: FAILED: MIGraphX is not within tolerance - check verbose output

     ✅ bert_large_uncased_fp16: PASSED: MIGraphX meets tolerance

     ✅ bert_large: PASSED: MIGraphX meets tolerance

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