reuse named tensor error message in generated code (#33536)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/33536
Simple fix, merge the identical string literals that were being inlined into every wrapper for ops that don't support named tensors. E.g.
```
Tensor all(const Tensor & self, int64_t dim, bool keepdim) {
if (self.has_names()) {
AT_ERROR(
"all is not yet supported with named tensors. Please drop names via "
"`tensor = tensor.rename(None)`, call the op with an unnamed tensor, "
"and set names on the result of the operation.");
}
const OptionalDeviceGuard device_guard(device_of(self));
return at::native::all(self, dim, keepdim);
}
```
becomes
```
Tensor all(const Tensor & self, int64_t dim, bool keepdim) {
if (self.has_names()) {
AT_ERROR("all", named_tensors_unsupported_error);
}
const OptionalDeviceGuard device_guard(device_of(self));
return at::native::all(self, dim, keepdim);
}
```
Also updated the generated file comments to include the source template names, e.g.
```
// generated by aten/src/ATen/gen.py from TypeDefault.cpp
```
Test Plan: Imported from OSS
Differential Revision: D19993407
Pulled By: bhosmer
fbshipit-source-id: 88395a649e6ba53191332344123555c217c5eb40
