Add default splits to MalNetTiny dataset

[flandolfi]

Hi,

I modified the MalNetTiny so that it now downloads also the official split information given by the authors. I also removed a previously added remove_isolated_nodes() from the process (added in #3592), that I believe it should be done as a transform/pre-transform (with RemoveIsolatedNodes), if needed by the user. This may be reverted, if you think otherwise!

The code now add the graphs not in alphabetical order, but in the order defined by the split files. One should be able to easily extend the MalNetTiny class to the full dataset MalNet just by modifying self.data_url, self.split_url, self.raw_file_names() and self.download(). I could not try it since I do not have enough space at the moment (the full data is about 130GB).

Francesco

[codecov]

Codecov Report

Merging #5078 (dd6db41) into master (a8f7a24) will decrease coverage by 1.89%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #5078      +/-   ##
==========================================
- Coverage   84.88%   82.98%   -1.90%     
==========================================
  Files         333      333              
  Lines       18375    18371       -4     
==========================================
- Hits        15598    15246     -352     
- Misses       2777     3125     +348     

Impacted Files	Coverage Δ
torch_geometric/nn/models/dimenet_utils.py	0.00% <0.00%> (-75.52%)	⬇️
torch_geometric/nn/models/dimenet.py	14.51% <0.00%> (-53.00%)	⬇️
torch_geometric/nn/conv/utils/typing.py	81.25% <0.00%> (-17.50%)	⬇️
torch_geometric/profile/profile.py	32.94% <0.00%> (-15.30%)	⬇️
torch_geometric/nn/inits.py	67.85% <0.00%> (-7.15%)	⬇️
torch_geometric/nn/resolver.py	88.00% <0.00%> (-6.00%)	⬇️
torch_geometric/transforms/add_self_loops.py	94.44% <0.00%> (-5.56%)	⬇️
torch_geometric/io/tu.py	93.90% <0.00%> (-2.44%)	⬇️
torch_geometric/transforms/virtual_node.py	97.56% <0.00%> (-2.44%)	⬇️
torch_geometric/nn/models/basic_gnn.py	89.71% <0.00%> (-1.72%)	⬇️
... and 8 more

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

[Padarn]

this looks good to me