SYNTHESIZER: Use only symbols from the original goto as terminals

[qinheping]

Collecting variables from trace will also collect those variables added during loop transformation, such as the car variables. Those variables should not appear in the loop contracts. This commit erases them from the terminals of the enumerating grammar used by the synthesizer.

• Each commit message has a non-empty body, explaining why the change was made.
• Methods or procedures I have added are documented, following the guidelines provided in CODING_STANDARD.md.
• The feature or user visible behaviour I have added or modified has been documented in the User Guide in doc/cprover-manual/
• Regression or unit tests are included, or existing tests cover the modified code (in this case I have detailed which ones those are in the commit message).
• My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[tautschnig]

Is it possible to include a test that demonstrates what difference this change makes?

[tautschnig]

Wouldn't it = result.erase(it); be sufficient here?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (2bab0b3) 78.83% compared to head (535bf95) 78.59%.
Report is 2 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #7970      +/-   ##
===========================================
- Coverage    78.83%   78.59%   -0.24%     
===========================================
  Files         1701     1701              
  Lines       196265   196270       +5     
===========================================
- Hits        154729   154264     -465     
- Misses       41536    42006     +470     

Files	Coverage Δ
...hesizer/enumerative_loop_contracts_synthesizer.cpp	96.34% <100.00%> (+0.06%)	⬆️
...nthesizer/enumerative_loop_contracts_synthesizer.h	100.00% <100.00%> (ø)

... and 45 files with indirect coverage changes

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

[qinheping]

Is it possible to include a test that demonstrates what difference this change makes?

The only difference is the performance of the enumeration. For example, it took 38 seconds to run the benchmark synthesizer_04
https://github.com/diffblue/cbmc/actions/runs/6506277751/job/17671527177?pr=7946#step:13:10913.
With this PR, the running time goes down to 2 seconds: https://github.com/diffblue/cbmc/actions/runs/6580619258/job/17878990179?pr=7970#step:13:10787.