Needed some features to generate the exports (needed for the cpp export refactor) :
Now possible to generate a label tensor into the generate_main_cpp() function for result verification at runtime.
Add a copy_folder() function and add a symlink argument to create symbolic links instead of simple file copy (help with dev process).
Add a function to remove the optional inputs from the order_inputs list so that they are not exported.
The ExportNode class now holds a static_folders variable acting as the existing static_files variable but for whole folders.
Change the kernels_to_copy export node system as the old system had some limitations. Indeed, it was not possible to chose a destination path different from the source path. Also the copied kernel would be automatically included in the fwd file, sometime leading to unused includes. The new system is based on a list of dict holding, for each kernel file, the source and destination paths as well as a boolean to indicate if the kernel needs to be included in the forward file.
Modified files
code_generation.py - Add copy_folder() function and symlink option.
export_registry.py - Add static_folder : list of folders to export.
scheduler_export.py - Calls the copy_folder() function to export the static_folders. Also change test_mode to dev_mode for more clarity. If the dev_mode is enabled, the copied files will be symbolic links to help with development.
generate_main.py - Add label export.
main.jinja - Add label export and change default folder for inputs/labels to ROOT/data.
main_compare.jinja - Change default folder for inputs/labels to ROOT/data.
export_utils.py - Add remove_optional_inputs() to remove the optional inputs of the graph before export.
