Add activation and dilation to conv1d

[sssshhhhhh]

• Optional activation for conv1d to fuse with bias
• Dilation support for conv1d gpu
• Use gemm_batch_strided to run each conv1d group in a single kernel

[jordimas]

Thanks. Have you done any benchmark on how performance is impacted?

[sssshhhhhh]

	bias	bias+gelu	none	bsz1
new	0.868059	0.869294	0.798557	0.20549
old	1.04657	1.10023	0.970503	0.205125

fp16 [4,1280,3000]*[1280,1280,3]=[4,1280,1500]
Same to a little faster

[sssshhhhhh]

Optimised im2col a bit, no bias/gelu down to 0.694081

[jordimas]

Thanks for your contribution