Forward Convolution Kernel

[m9ssah]

Hello, I'm interested in helping implement the forward convolution kernel, which is marked as 📅 Planned in ROADMAP.md.
After browsing through the repository, I've come up with a couple of questions/thoughts to shape the implementation direction

Design proposal:

1. Following the existing convention that op signatures mirror PyTorch's, I plan on implementing separate variants, such as conv1d, conv2d, etc., as opposed to one master conv op
2. I think starting with conv1d first would be more effective. Perhaps I can create a draft PR so that I can receive feedback before expanding the scope for larger dimensions
3. bf16/fp16 inputs with FP32 accumulation. FP32 inputs via TF32 on tensor cores to match matmul.py
4. NCHW layout
5. Benchmark against F.convNd (cuDNN)

Questions:

1. Are there any internal blockers or overlapping work not yet reflected in the repository that I should be aware of?
2. Is an NHCW eventually needed or is NCHW sufficient?
3. I am currently using an RTX 5060 Ti, meaning that if I conduct performance profiling before making a PR, my numbers likely won't be as decisive compared to a B200 card. Is that ok?

Any additional tips/help/feedback would be much appreciated. Thank you! :D