compiler: Tweak caching in fusion pass, along with DAG construction

[EdCaunt]

Provides substantial speedup in the fusion pass through a mixture of tweaks to caching and improved DAG construction using transitive reduction. Stacked with some other branches I have, I was seeing 1.5x - 3x compilation speedup for CPML-like operators

This is entirely Claude/Codex to begin with so will need some review for sanity checking and human readability.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 84.32%. Comparing base (0164e85) to head (9b4c1d3).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3051      +/-   ##
==========================================
+ Coverage   84.26%   84.32%   +0.05%     
==========================================
  Files         258      258              
  Lines       56982    57153     +171     
  Branches     4854     4875      +21     
==========================================
+ Hits        48016    48194     +178     
+ Misses       8139     8134       -5     
+ Partials      827      825       -2     

Flag	Coverage Δ
pytest-gpu-aomp-amdgpuX	68.59% <96.22%> (+0.03%)	⬆️
pytest-gpu-gcc-	79.11% <100.00%> (+0.07%)	⬆️
pytest-gpu-icx-	79.04% <100.00%> (+0.06%)	⬆️
pytest-gpu-nvc-nvidiaX	69.17% <96.22%> (+0.04%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.