Modify the implementation of sagpool so it is consistent with the description in paper

[ChenYizhu97]

The current implementation of sagpool reuses the SelectTopK class, which introduces an extra learnable parameter that learns to take the negative value of the attention score or not.

The top k nodes should be selected by their attentions score attn, but in current implementation, the top k nodes are selected by attn*i/norm(i), where i is the extra learnable parameter. The value of i/norm(i) is 1 if i >0, -1 if i<0. That means when i<0 the current select function will return bottom k nodes.

[codecov]

Codecov Report

Attention: Patch coverage is 92.85714% with 2 lines in your changes missing coverage. Please review.

Project coverage is 88.72%. Comparing base (90e7604) to head (cb9a9a1).
Report is 411 commits behind head on master.

Files with missing lines	Patch %	Lines
torch_geometric/nn/pool/select/sag.py	92.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #8562      +/-   ##
==========================================
- Coverage   88.73%   88.72%   -0.02%     
==========================================
  Files         479      480       +1     
  Lines       29847    29872      +25     
==========================================
+ Hits        26486    26504      +18     
- Misses       3361     3368       +7     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.