Apply ConsecutiveSlice optimizations to CompiledSimulation for ~46% reduction in compile time.

[fdxmw]

For pyrtlnet, this reduces CompiledSimulation's startup time from 2.96 seconds to 1.6 seconds. It does not affect simulation speed, so the compiler was optimizing the inefficient code we were generating before, but we were paying for those optimizations with startup time.

Also: Simplify the interface for make_consecutive_slices and move the shift helper in simulation.py to toplevel so we can call it from compilesim.py. Add some comments and emit bitmasks in hex so they're easier to identify.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.2%. Comparing base (1c9ce1f) to head (065d98e).

Additional details and impacted files

@@              Coverage Diff              @@
##           development    #495     +/-   ##
=============================================
+ Coverage         92.2%   92.2%   +0.1%     
=============================================
  Files               30      30             
  Lines             7465    7488     +23     
=============================================
+ Hits              6876    6902     +26     
+ Misses             589     586      -3     

☔ 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.

[fdxmw]

Example: the old code would generate:

((1&(w4134_layer1matmulreg10[0]>>0))<<0)|
 ((1&(w4134_layer1matmulreg10[0]>>1))<<1)|
 ((1&(w4134_layer1matmulreg10[0]>>2))<<2)|
 ((1&(w4134_layer1matmulreg10[0]>>3))<<3)|
 ((1&(w4134_layer1matmulreg10[0]>>4))<<4)|
 ((1&(w4134_layer1matmulreg10[0]>>5))<<5)|
 ((1&(w4134_layer1matmulreg10[0]>>6))<<6)|
 ((1&(w4134_layer1matmulreg10[0]>>7))<<7)

while the new code generates:

(w4134_layer1matmulreg10[0] & 0xFF)