Use the float8 and generic_float template for specialization

[pfultz2]

This is uses the template for float8 so we no longer need to add so many combinations of specializations. This should cover all of them. Also this will avoid the errors when merging #3570 and #3578 together(which will also cause an explosion of specializations).

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.17%. Comparing base (f5df004) to head (eaf728d).
Report is 5 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3606      +/-   ##
===========================================
- Coverage    92.17%   92.17%   -0.01%     
===========================================
  Files          513      513              
  Lines        21536    21533       -3     
===========================================
- Hits         19851    19848       -3     
  Misses        1685     1685              

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

[migraphx-bot]

Test	Batch	Rate new eaf728	Rate old f5df00	Diff	Compare
torchvision-resnet50	64	3,257.73	3,258.22	-0.02%	✅
torchvision-resnet50_fp16	64	6,987.12	6,994.86	-0.11%	✅
torchvision-densenet121	32	2,435.03	2,438.08	-0.12%	✅
torchvision-densenet121_fp16	32	4,101.95	4,082.99	0.46%	✅
torchvision-inceptionv3	32	1,637.85	1,637.00	0.05%	✅
torchvision-inceptionv3_fp16	32	2,762.10	2,762.55	-0.02%	✅
cadene-inceptionv4	16	776.33	776.12	0.03%	✅
cadene-resnext64x4	16	808.08	811.97	-0.48%	✅
slim-mobilenet	64	7,533.07	7,534.11	-0.01%	✅
slim-nasnetalarge	64	211.49	211.46	0.02%	✅
slim-resnet50v2	64	3,504.30	3,503.66	0.02%	✅
bert-mrpc-onnx	8	1,149.63	1,150.99	-0.12%	✅
bert-mrpc-tf	1	468.39	494.60	-5.30%	🔴
pytorch-examples-wlang-gru	1	418.13	428.98	-2.53%	✅
pytorch-examples-wlang-lstm	1	389.52	385.66	1.00%	✅
torchvision-resnet50_1	1	800.42	766.70	4.40%	🔆
cadene-dpn92_1	1	400.66	433.66	-7.61%	🔴
cadene-resnext101_1	1	383.65	383.99	-0.09%	✅
onnx-taau-downsample	1	341.78	343.15	-0.40%	✅
dlrm-criteoterabyte	1	33.34	33.33	0.03%	✅
dlrm-criteoterabyte_fp16	1	52.70	52.72	-0.04%	✅
agentmodel	1	7,758.07	8,297.10	-6.50%	🔴
unet_fp16	2	58.87	58.77	0.17%	✅
resnet50v1_fp16	1	964.74	948.38	1.72%	✅
resnet50v1_int8	1	1,022.74	1,046.77	-2.30%	✅
bert_base_cased_fp16	64	1,171.28	1,171.13	0.01%	✅
bert_large_uncased_fp16	32	363.50	363.68	-0.05%	✅
bert_large_fp16	1	200.57	197.66	1.47%	✅
distilgpt2_fp16	16	2,203.34	2,203.57	-0.01%	✅
yolov5s	1	550.17	549.13	0.19%	✅
tinyllama	1	43.43	43.43	0.01%	✅
vicuna-fastchat	1	174.11	169.54	2.69%	✅
whisper-tiny-encoder	1	418.81	418.17	0.15%	✅
whisper-tiny-decoder	1	428.95	428.69	0.06%	✅

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

[CharlieL7]

LGTM, I make changes to #3570 to match.

[richagadgil]

Will adjust BF16 PR (#3578) accordingly.