adding gen_archive optional parameter description

Author: VGPReys

src/global_parameters.md

@@ -34,14 +34,15 @@ Various parameters are related to the execution modes:
 - `offline`: When set to `true`, completely isolate the haddock3 run and results from internet. This option is useful when no internet connection is available. Default is `false`.
 - `debug`: By setting it to `false`, reduces the amount of I/O operations, often speeding up the process. When set to `true`, input files, intermediate files and output files are generated and kept, which is useful when tracking potential errors. The default is `false`.
 
-### Local mode
+
+#### Local mode
 
 Often the prefered execution mode if you submit a haddock3 run to a queuing system or run on your own computer.
 The `local` mode (targeted using the global parameter `mode = 'local'`), utilize the operating system device to perform the computations.
 Setting the `ncores` parameter allows to tune the number of CPU cores to use during the run.
 Note that if you set this value too high compared to your system capabilities, this value will be automatically scaled down to the maximum number of cores available on the machine.
 
-### Batch mode
+#### Batch mode
 
 Utilise queuing system machinery to submit CNS runs.
 
@@ -50,11 +51,12 @@ Utilise queuing system machinery to submit CNS runs.
 - `queue_limit`: Sets the number of jobs to submit to the batch system. Default is 100.
 - `concat`: Number of models to produce per job to send to the batch system. If set to a value above 1, multiple models can be calculated within one job. The default is 1.
 
-### MPI mode
+#### MPI mode
 
 Requires the installation of the `mpi4py` python library and `OpenMPI` to be installed on the operating system.
 
 ## Optional global parameters
 
 - `postprocess`: When set to `true`, executes `haddock3-analyse` on the CAPRI folders at the end of the run. The default is `true`.
 - `preprocess`: When set to `true`, tries to correct input PDBs before the workflow. The default is `false`.
+- `gen_archive`: By setting it to `true`, at the end of the workflow, the entire `run_dir` will be compressed as a `.tgz` archive. In addition, the `analysis` directory will be archived too, and a local copy of models will be made in it. This allows to reduce the number of files to be transferred and make the analysis directory self-sufficient. Useful for the web application and if only the last models generated by the workflow are required. The default is `false`.