Add OCP FP8 formats

[CharlieL7]

• Enables/adds the F8E4M3FN and F8E5M2 data types
• Change to gpu::context value to be able to get the gfx name at quantization
• Update tests to also try the new data types
• rocBLAS does not support the OCP fp8 formats so converting the datatype before rocBLAS instructions
  • hipBLASlt should have support and will need an update to the code later
• See Update MIGraphX FP8 to support OCP types #3393 for issue details
• Needs Handle the new F8 types in RockTuningImpl.cpp too. Oops. rocMLIR#1634 from rocMLIR to work fully on Navi4X

[codecov]

Codecov Report

Attention: Patch coverage is 50.00000% with 11 lines in your changes missing coverage. Please review.

Project coverage is 92.04%. Comparing base (8ff42e8) to head (5451f91).
Report is 164 commits behind head on develop.

Files with missing lines	Patch %	Lines
src/quantization.cpp	18.18%	9 Missing ⚠️
src/include/migraphx/op/quant_convolution.hpp	50.00%	1 Missing ⚠️
src/include/migraphx/op/quant_dot.hpp	75.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3399      +/-   ##
===========================================
- Coverage    92.07%   92.04%   -0.04%     
===========================================
  Files          504      505       +1     
  Lines        20689    20699      +10     
===========================================
+ Hits         19050    19052       +2     
- Misses        1639     1647       +8     

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

[TedThemistokleous]

Make the TODOs separate issues so we can track them.

I kicked off CI for the other build checks. Just ensure CI is working and you add tickets and I'll approve.

[TedThemistokleous]

CI failure seems unrelated to your changes. Getting a lot of "Failed to fetch" errors.

[TedThemistokleous]

Approved.. I'll rekick CI but that MLIR issue seems like its server related and not build.

[migraphx-bot]

Test	Batch	Rate new 5451f9	Rate old b87355	Diff	Compare
torchvision-resnet50	64	3,249.70	3,251.19	-0.05%	✅
torchvision-resnet50_fp16	64	6,989.55	6,988.29	0.02%	✅
torchvision-densenet121	32	2,432.87	2,433.72	-0.03%	✅
torchvision-densenet121_fp16	32	4,105.26	4,109.85	-0.11%	✅
torchvision-inceptionv3	32	1,635.92	1,636.15	-0.01%	✅
torchvision-inceptionv3_fp16	32	2,739.66	2,740.01	-0.01%	✅
cadene-inceptionv4	16	776.75	776.10	0.08%	✅
cadene-resnext64x4	16	807.38	808.31	-0.12%	✅
slim-mobilenet	64	7,453.43	7,462.97	-0.13%	✅
slim-nasnetalarge	64	208.28	208.06	0.10%	✅
slim-resnet50v2	64	3,435.76	3,344.04	2.74%	✅
bert-mrpc-onnx	8	1,154.16	1,149.25	0.43%	✅
bert-mrpc-tf	1	309.90	307.92	0.64%	✅
pytorch-examples-wlang-gru	1	430.90	414.16	4.04%	🔆
pytorch-examples-wlang-lstm	1	383.72	376.68	1.87%	✅
torchvision-resnet50_1	1	801.78	814.48	-1.56%	✅
cadene-dpn92_1	1	395.99	400.81	-1.20%	✅
cadene-resnext101_1	1	383.40	380.42	0.78%	✅
onnx-taau-downsample	1	344.90	344.67	0.07%	✅
dlrm-criteoterabyte	1	35.05	35.06	-0.04%	✅
dlrm-criteoterabyte_fp16	1	58.12	58.10	0.02%	✅
agentmodel	1	8,081.40	8,179.29	-1.20%	✅
unet_fp16	2	58.03	57.98	0.08%	✅
resnet50v1_fp16	1	932.44	985.87	-5.42%	🔴
resnet50v1_int8	1	949.16	935.52	1.46%	✅
bert_base_cased_fp16	64	1,154.07	1,154.71	-0.05%	✅
bert_large_uncased_fp16	32	355.86	355.97	-0.03%	✅
bert_large_fp16	1	211.27	211.90	-0.30%	✅
distilgpt2_fp16	16	2,162.77	2,159.23	0.16%	✅
yolov5s	1	530.35	540.08	-1.80%	✅
tinyllama	1	43.40	43.46	-0.13%	✅
vicuna-fastchat	1	170.26	179.52	-5.16%	🔴
whisper-tiny-encoder	1	417.83	418.25	-0.10%	✅
whisper-tiny-decoder	1	433.46	433.10	0.08%	✅

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