#5629: fail with ExFailure on scanf format ending in dangling %

[morphqdd]

Closes #5629.

EOscanf.lambda() builds the match regex in a first loop, where a trailing lone % sets literal = true and appends nothing, without raising any error for the dangling, never-completed specifier. The extraction loop then locates that % via frmt.indexOf('%') and reads the following char with frmt.charAt(idx + 1). When the % is the last character of the format, idx + 1 == frmt.length(), and a raw java.lang.StringIndexOutOfBoundsException escapes instead of a proper ExFailure.

This adds a check right after the first loop: if literal is still true once the loop ends, the format string has a dangling % with no specifier after it, so it now fails fast with a clear ExFailure, matching the existing pattern used for an unsupported specifier (the default: branch a few lines above).

Added EOscanfTest (a Java unit test asserting ExFailure is thrown for scanf "hello%" "hello") and a --> on-scanf-with-dangling-percent EO-level example in scanf.eo, next to the two existing on-scanf-with-* negative examples for this object.

Verified locally: the new tests pass, EOstringTest (51 tests) and the generated EOscanfEOAtomTest (15 tests, including the new example) pass, and qulice is clean.

[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 Comparison Unavailable

Unfortunately, one of the benchmarks is missing, and we couldn't generate a performance comparison report.

Please ensure that both the base and PR benchmark results are available for analysis.

[0crat]

@morphqdd Thanks for the contribution! You've earned +16 points. Please, keep them coming. Your running score is +1274; don't forget to check your Zerocracy account too).