[x] added chunking and get_chunks to all algorithms(in my gsoc proposal this was an "expanding" goal, not a "revisiting" goal, but it seemed more apt to include it here)
[x] testing all get_chunks together(rn ignores funcs that require args other than G)
[x] enhanced is_reachable and tournament_is_strongly_connected in tournament.py(can be experimented with and enhanced further but probably should leave that for a different PR)
[x] renamed the Dispatcher class to BackendInterface (ref. comment) and updated it so that running nxp.ParallelGraph() or nxp.ParallelGraph([(1, 2), (2, 3)]) doesn't give any error and added __str__ so that ParallelGraph with n nodes and m edges is printed instead of <nx_parallel.interface.ParallelGraph object at 0x10495e6d0> when a ParallelGraph is passed in print() (This is in experimental section of the proposal)
get_chunksto all algorithms(in my gsoc proposal this was an "expanding" goal, not a "revisiting" goal, but it seemed more apt to include it here)get_chunkstogether(rn ignores funcs that require args other than G)is_reachableandtournament_is_strongly_connectedintournament.py(can be experimented with and enhanced further but probably should leave that for a different PR)pip installs intest.ymlin gh wfDispatcherclass toBackendInterface(ref. comment) and updated it so that runningnxp.ParallelGraph()ornxp.ParallelGraph([(1, 2), (2, 3)])doesn't give any error and added__str__so thatParallelGraph with n nodes and m edgesis printed instead of<nx_parallel.interface.ParallelGraph object at 0x10495e6d0>when aParallelGraphis passed inprint()(This is in experimental section of the proposal)