bit_cast operator

[CharlieL7]

• Converts between types while keeping the data the same. Basically an extension of reinterpret_cast without undefined behavior
• Needed for conversion pass between fp8e4m3fn to fp8e4m3fnuz.

[pfultz2]

Call migraphx::bit_cast function instead of using the builtin.

[migraphx-bot]

Test	Batch	Rate new a15e5a	Rate old 0ad807	Diff	Compare
torchvision-resnet50	64	3,256.68	3,239.44	0.53%	✅
torchvision-resnet50_fp16	64	6,981.94	6,989.98	-0.12%	✅
torchvision-densenet121	32	2,432.96	2,435.40	-0.10%	✅
torchvision-densenet121_fp16	32	4,075.78	4,083.28	-0.18%	✅
torchvision-inceptionv3	32	1,628.87	1,630.36	-0.09%	✅
torchvision-inceptionv3_fp16	32	2,747.37	2,743.98	0.12%	✅
cadene-inceptionv4	16	765.37	765.72	-0.04%	✅
cadene-resnext64x4	16	811.27	811.01	0.03%	✅
slim-mobilenet	64	7,465.93	7,471.20	-0.07%	✅
slim-nasnetalarge	64	208.54	208.50	0.02%	✅
slim-resnet50v2	64	3,443.82	3,438.88	0.14%	✅
bert-mrpc-onnx	8	1,150.80	1,147.12	0.32%	✅
bert-mrpc-tf	1	463.69	469.98	-1.34%	✅
pytorch-examples-wlang-gru	1	420.31	416.34	0.95%	✅
pytorch-examples-wlang-lstm	1	388.84	463.26	-16.07%	🔴
torchvision-resnet50_1	1	766.24	770.23	-0.52%	✅
cadene-dpn92_1	1	400.58	399.89	0.17%	✅
cadene-resnext101_1	1	382.51	382.62	-0.03%	✅
onnx-taau-downsample	1	346.05	345.09	0.28%	✅
dlrm-criteoterabyte	1	33.36	33.34	0.05%	✅
dlrm-criteoterabyte_fp16	1	52.72	52.72	-0.00%	✅
agentmodel	1	8,248.51	10,261.49	-19.62%	🔴
unet_fp16	2	58.84	58.87	-0.05%	✅
resnet50v1_fp16	1	934.60	982.51	-4.88%	🔴
resnet50v1_int8	1	997.90	1,007.17	-0.92%	✅
bert_base_cased_fp16	64	1,170.20	1,169.34	0.07%	✅
bert_large_uncased_fp16	32	362.87	363.46	-0.16%	✅
bert_large_fp16	1	198.76	198.15	0.31%	✅
distilgpt2_fp16	16	2,201.93	2,201.47	0.02%	✅
yolov5s	1	531.87	536.48	-0.86%	✅
tinyllama	1	43.46	43.38	0.19%	✅
vicuna-fastchat	1	174.24	172.97	0.73%	✅
whisper-tiny-encoder	1	418.79	418.04	0.18%	✅
whisper-tiny-decoder	1	428.48	423.83	1.10%	✅

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