Remove unused device kernels

[umangyadav]

Gather and Pad operators have JIT kernels.
Device kernels are no longer in use hence removing them.

[TedThemistokleous]

I get cleaning these up but do we ever want to keep this around for debug purposes?

[umangyadav]

I get cleaning these up but do we ever want to keep this around for debug purposes?

I see value in that but we can use git to retrieve older files. Also we had gather and pad JIT kernels for a while now. So unlikely it has any bugs now.
I am working on FP8 for device kernels. Having these files that are unused have added burden to keep it maintained otherwise.

[migraphx-bot]

Test	Batch	Rate new 09e857	Rate old 02f740	Diff	Compare
torchvision-resnet50	64	2,830.66	2,832.03	-0.05%	✅
torchvision-resnet50_fp16	64	6,501.27	6,501.16	0.00%	✅
torchvision-densenet121	32	2,091.38	2,098.99	-0.36%	✅
torchvision-densenet121_fp16	32	3,650.81	3,666.47	-0.43%	✅
torchvision-inceptionv3	32	1,585.85	1,584.68	0.07%	✅
torchvision-inceptionv3_fp16	32	2,569.46	2,569.37	0.00%	✅
cadene-inceptionv4	16	704.54	704.57	-0.00%	✅
cadene-resnext64x4	16	692.72	691.97	0.11%	✅
slim-mobilenet	64	8,336.64	8,324.07	0.15%	✅
slim-nasnetalarge	64	225.25	225.67	-0.18%	✅
slim-resnet50v2	64	2,663.85	2,665.32	-0.06%	✅
bert-mrpc-onnx	8	823.32	822.26	0.13%	✅
bert-mrpc-tf	1	389.15	386.92	0.58%	✅
pytorch-examples-wlang-gru	1	298.57	300.38	-0.60%	✅
pytorch-examples-wlang-lstm	1	312.48	310.78	0.55%	✅
torchvision-resnet50_1	1	603.36	598.68	0.78%	✅
torchvision-inceptionv3_1	1	342.17	343.19	-0.30%	✅
cadene-dpn92_1	1	398.72	401.44	-0.68%	✅
cadene-resnext101_1	1	328.49	324.52	1.22%	✅
slim-vgg16_1	1	459.63	459.52	0.02%	✅
slim-mobilenet_1	1	2,099.23	2,116.32	-0.81%	✅
slim-inceptionv4_1	1	218.38	220.17	-0.81%	✅
onnx-taau-downsample	1	304.65	304.43	0.07%	✅
dlrm-criteoterabyte	1	21.60	21.60	0.00%	✅
dlrm-criteoterabyte_fp16	1	40.65	40.66	-0.04%	✅
agentmodel	1	6,010.23	5,930.88	1.34%	✅
unet_fp16	2	54.75	54.75	0.01%	✅
resnet50v1_fp16	1	937.86	959.61	-2.27%	✅
bert_base_cased_fp16	64	903.03	903.28	-0.03%	✅
bert_large_uncased_fp16	32	285.75	285.65	0.04%	✅
bert_large_fp16	1	166.47	166.48	-0.01%	✅
distilgpt2_fp16	16	1,280.74	1,282.20	-0.11%	✅

This build is OK for 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