Value-set based dereferencing: fix simplified handling of *(p + i)

[tautschnig]

Value-set based dereferencing must not take an access path through an object that precludes a subsequent index expression from accessing a different part of the object. Such a situation can arise when the value set has a known (constant) offset for the pointer that would identify one particular element in an array (within that object). The code using value-set based dereferencing, however, may be trying to resolve a subexpression of an index expression, where said index expression would lead to a different element that may itself be part of a different array within the same overall object.

Fixes: #8570

• 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.
• n/a 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).
• n/a My commit message includes data points confirming performance improvements (if claimed).
• My PR is restricted to a single feature or bugfix.
• n/a White-space or formatting changes outside the feature-related changed lines are in commits of their own.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.65%. Comparing base (66004dc) to head (164407f).
Report is 15 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #8578      +/-   ##
===========================================
- Coverage    78.71%   78.65%   -0.06%     
===========================================
  Files         1732     1732              
  Lines       199536   199673     +137     
  Branches     18281    18244      -37     
===========================================
- Hits        157057   157056       -1     
- Misses       42479    42617     +138     

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

[remi-delmas-3000]

The commit adds a new argument recurse_into_arrays to the get_subexpression_at_offset to force fallback to using byte-extracts but I could not really understand what condition triggers the fallback ? Is the idea to try to resolve a lookup expression as a cell/field access if enough info is known about the index and default to a flat/unstructured by-extract otherwise ?

[tautschnig]

@kroening Reworked as discussed.