[codex] Fix LINE edge sampling probabilities

[shenweichen]

Summary

• Normalize LINE edge sampling probabilities before passing them to the alias table.
• Add a regression test that verifies both node and edge sampling tables receive probability distributions summing to 1.

Root cause

create_alias_table expects normalized probabilities and scales them internally by table length. LINE edge sampling already multiplied each edge weight by numEdges, so the alias input summed to the number of edges instead of 1.

Addresses #31 and #51. This may also help investigate the LINE NaN reports in #33 and #73.

Checks

• git diff --check
• python -m pytest tests/line_test.py
• python -m pytest tests/examples_test.py -k line

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 0 complexity · 0 duplication

Metric	Results
Complexity	0
Duplication	0

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}