Merged the elements that differed in optados_mod_working by hand to
ensure functionality was retained. Untested!

Changed minor things in cell.f90 to allow the photoemission to work

The Calculation of epsilon_2 is cell volume dependent. This was changed,
so that epsilon_2 uses photo_slab_volume, which should lessen the
cell volume dependence of the QE values.
Preliminary data suggests otherwise, so this is a WIP

Merged 'Incorrectly Commited Changes into Development Branch'

Changed the condition from equal to zero to less than tiny number for
what kind of formula is used to compute the ARPES Phi angle

Implementing the use of user supplied imfp values for mixed material
systems

Added statements to use the correct volume (i.e. photo_slab_volume in
the case dos_per_volume during a photoemission run is enabled.

Added the option to turn off the extra output files and made it the
default behaviour.

Code in dos_utils.f90 checked for uninitialised variables since they
belong to other smearing methods. Changed the default efermi setting to
rectify the parameters printing function.
Changed printing in photo.f90 to make function calls clearer

Fixed more occurences where 'if' conditions only applicable with certain
parameters were always evaluated, leading to denormal and underflow
exceptions. Changed some parameter printing output to be more in line with
the other output

The output and input for simultaneous calculations in a single folder
is now implemented as an option specified as an argument at runtime
rather than a parameter present in the .odi file.

Added printing to extra file to aid debugging

…o_dev into develop

Tested sweep, but single energy version is untested - WIP

Successful run of single photon energy implementation

Reworked the  structure of the code:
* checked for non deallocated arrays, of which there were quite a few
* changed the position of quite a few deallocation statements for more
efficient memory usage
* changed the order of calling subroutines to reflect the general order
of operations needed for QE calculations
* inserted if statements to only run certain subroutines if needed
* removed extra variables if it was more efficient to do the operations
in place on the final variable (array)

Implemented calls to deallocation subroutines to reduce amount of lost
memory references. Included deallocation statements where necessary

To remove div by 0 error the subroutine make_foptical_weights had to be
put in a different place
Added another deallocation statement

Removed extra deallocation subroutine

…o_dev into develop

routines

Replaced a do loop with a sum and if the occupancy of a photoemission
final state -is = 0, then skip

The indices and loops were constructed in a way, that a lot of random
reads and writes were made. This was tried to change without touching
other modules

Put the foptical_matrix_weights subroutine in the approximate location
within the file, where it is in the call graph.

Added the if statement optimisation to the jdos subroutine with a
disclaimer and some bit of code, that checks the bands are ordered by
energy and stops the calculation if that is the case.

Implemented same ordering of loops in the one step calculation sub-
routine, as the three step subroutine.

The parameter photoemission clashed with photoemission_sweep so it was
renamed to photon_sweep again

More: cleaned up the photoemission output

Replaced the multi output with a temporary folder printing functionality
to allow for faster local printing and copying after the calculation has
finished.

The implementation of MPI parallelisation for the photoemission module
gives a compiling code, that can run fully without encountering errors.
Calculated values of photoemission are wrong!!