[quantization] Dispatch key error message is confusing to users #28518
Description
For example: #28483
User encounters the following inscrutable error message:
RuntimeError: Didn't find kernel to dispatch to for operator 'quantized::conv2d'. Tried to look up kernel for dispatch key 'CPUTensorId'. Registered dispatch
keys are: [QuantizedCPUTensorId] (lookup_ at /opt/conda/conda-bld/pytorch_1570910687650/work/aten/src/ATen/core/dispatch/DispatchTable.h:249)
We should figure out a way to print a more friendly error message, such as:
Tried to run quantized::conv2d, but the input was not quantized! Please ensure you have QuantStub during model conversion, or you explicitly call one of quantize_per_tensor, quantize_per_channel