Updating run script for ruche gpu ci

Update anaconda module for ruche

Updated gpu compilation ruche CI submition script

Added support for cuda >= 13.0 ; for now cuda < 13.0 and hip are doing the same thing, will need to be updated if newer hip follows newer cuda ; had to switch to c++17 standard ; support for the new ruche stack and cleanup old machinefiles

Some doc in order to using new smoothed beam laser profil in 2D, 3D, 2D periodic, 3D periodic

Fixed sumAllComponents and removed a std::cout

Some laser profile in order to make Temporal (broadband,ssd) and Spatial Smoothed beam (with speckles)

Merge branch 'updated-AM-BC' into 'develop'

Merge branch 'develop' into updated-AM-BC

Merge branch 'develop' of llrgit.in2p3.fr:smilei/smilei into develop

clarification in the AMcylindrical geometry documentation

Merge branch 'develop' into updated-AM-BC

Merge branch 'GPU_4th_order' into 'develop'

Merge branch 'develop' into GPU_4th_order

added copyFieldsFromDeviceToHost back to the moving window function in order to fix MW for multi GPU execution ; changed HostDeviceMemoryManagement::DoGetDevicePointer to have the same behavior for openacc and openmp , requires openmp > 5.1 which is standard now