EnergyToFluence Refactoring + Better SystemLevelTests

The SystemLevelTests (tcl and python) now use the normalized L1Norm of errors to determine if their results are within the error margin of the gold standard results. Removed the Debug version of the tests since the error comparsions are not accurate enough for the smaller amount of packets that are being simulated. Refactored the EnergyToFluence class: materials() and geometry() functions have been replaced by the kernel() function. The kernel already incorporates all required information and reducing the amount of redundant code that needs to be written to setup the conversion. Added a new function called energy(). This function allows to specifiy how much energy the number of simulated packets represent thus making the reuslts clearer in a physical sense for the users. Added a new input and output function to describe the type of input/output that is wanted by the conversion class. Now we explicitly have PhotonWeight as input to the conversion and everything else will be scaled to actual energy values if they are provided. Updated all reg tests that were impacted by these changes. Implemendted some minor fixes for the python interface where some functions could not be wrapped because of conflicting integrated python functions