Add case check for external dependency classes

[Jeckerson]

Hello!

In raising this pull request, I confirm the following:

• I have checked that another pull request for this purpose does not exist
• I wrote some tests for this PR
• I updated the CHANGELOG

Review of #2499 and different vision approach.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[github-actions]

Benchmark report

Comparison against development (7cad41fe7dd8e803608112dd0407ad0b69c07976) on fix-external-class-case-sensitive (6d7a14424e98fb981ee9808bcbe30f250dbde15e).
Each row's mode column shows the head-branch absolute throughput and the percent delta vs base. Positive deltas on Zephir* subjects mean head is faster; deltas on Php* baseline subjects are noise-floor signal.

+---------------------+----------------------------------------+-----+------+-----+----------------+-----------------------+------------------+
| benchmark           | subject                                | set | revs | its | mem_peak       | mode                  | rstdev           |
+---------------------+----------------------------------------+-----+------+-----+----------------+-----------------------+------------------+
| PropertyAccessBench | benchZephirPropertyRead                |     | 500  | 5   | 2.270mb +0.01% | 0.289779ops/μs +0.72% | ±2.43% +35.57%   |
| PropertyAccessBench | benchPhpPropertyRead                   |     | 500  | 5   | 2.270mb +0.01% | 0.665255ops/μs -3.33% | ±1.49% -81.24%   |
| StringConcatBench   | benchZephirStringConcat                |     | 500  | 5   | 2.270mb +0.01% | 0.948668ops/μs -0.82% | ±2.62% +44.93%   |
| StringConcatBench   | benchPhpStringConcat                   |     | 500  | 5   | 2.270mb +0.01% | 0.722274ops/μs -0.08% | ±2.02% -52.24%   |
| NewInstanceBench    | benchZephirNewInstanceLoop             |     | 200  | 5   | 2.270mb +0.01% | 0.079438ops/μs -0.98% | ±1.12% -55.65%   |
| NewInstanceBench    | benchPhpNewInstanceLoop                |     | 200  | 5   | 2.270mb +0.01% | 0.342044ops/μs +0.55% | ±11.18% +389.26% |
| ArrayAccessBench    | benchZephirArrayFetchString            |     | 500  | 5   | 2.270mb +0.01% | 0.345915ops/μs -3.55% | ±1.00% +6.98%    |
| ArrayAccessBench    | benchPhpArrayFetchString               |     | 500  | 5   | 2.270mb +0.01% | 0.363147ops/μs -0.75% | ±3.15% +130.26%  |
| MethodCallBench     | benchZephirMethodDispatch              |     | 500  | 5   | 2.270mb +0.01% | 0.109698ops/μs -0.99% | ±0.87% +6.54%    |
| MethodCallBench     | benchPhpMethodDispatch                 |     | 500  | 5   | 2.270mb +0.01% | 0.456760ops/μs +0.09% | ±3.16% +366.74%  |
| VariadicBench       | benchZephirVariadicNarrow              |     | 1000 | 5   | 2.270mb +0.01% | 0.048614ops/μs +0.68% | ±0.19% -49.05%   |
| VariadicBench       | benchZephirVariadicWide                |     | 1000 | 5   | 2.270mb +0.01% | 0.031402ops/μs +1.50% | ±1.02% +45.37%   |
| VariadicBench       | benchZephirFuncGetArgs                 |     | 1000 | 5   | 2.270mb +0.01% | 0.044871ops/μs +3.89% | ±0.39% -25.94%   |
| VariadicBench       | benchPhpVariadic                       |     | 1000 | 5   | 2.270mb +0.01% | 0.070134ops/μs +0.30% | ±1.37% +227.64%  |
| ForLoopBench        | benchZephirForOverArray                |     | 1000 | 5   | 2.270mb +0.01% | 5.912573ops/μs +5.50% | ±12.03% +243.94% |
| ForLoopBench        | benchPhpForOverArray                   |     | 1000 | 5   | 2.270mb +0.01% | 1.681845ops/μs +4.11% | ±3.60% +31.44%   |
| ForLoopBench        | benchZephirForOverIterator             |     | 1000 | 5   | 2.270mb +0.01% | 0.291185ops/μs -2.96% | ±0.53% -37.98%   |
| ForLoopBench        | benchPhpForOverIterator                |     | 1000 | 5   | 2.270mb +0.01% | 1.996616ops/μs -1.39% | ±4.58% +136.84%  |
| ForLoopBench        | benchZephirForOverIteratorWithContinue |     | 1000 | 5   | 2.270mb +0.01% | 0.287954ops/μs -3.02% | ±0.63% -29.86%   |
| ForLoopBench        | benchPhpForOverIteratorWithContinue    |     | 1000 | 5   | 2.270mb +0.01% | 1.826272ops/μs +1.18% | ±1.79% +32.02%   |
+---------------------+----------------------------------------+-----+------+-----+----------------+-----------------------+------------------+

PHP 8.3 on Linux (X64). Micro-benchmarks are noisy on shared runners (±5-20% per subject); treat any single-digit delta as inconclusive.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 19.45%. Comparing base (7cad41f) to head (6d7a144).

Additional details and impacted files

@@               Coverage Diff               @@
##           development    #2556      +/-   ##
===============================================
+ Coverage        19.29%   19.45%   +0.16%     
===============================================
  Files              288      288              
  Lines            24501    24504       +3     
===============================================
+ Hits              4728     4768      +40     
+ Misses           19773    19736      -37     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.