fix(plan_generation): properly parse LLM validation response instead of bypassing on code length

The _validate method previously used `or len(all_code) > 10` as a fallback,
which caused any code longer than 10 characters to auto-pass regardless of
the LLM's actual assessment. This made LLM validation completely ineffective.

The fix removes the code-length bypass and now correctly determines
validity based solely on whether the LLM explicitly responds with PASS or FAIL.

Fixes #10480

src/cleveragents/agents/graphs/plan_generation.py

@@ -527,8 +527,8 @@ class PlanGenerationGraph:
             )
             validation = str(result)
 
-            # Simple validation check (in real implementation, parse the LLM response)
-            is_valid = "PASS" in validation.upper() or len(all_code) > 10
+            # Check whether the LLM explicitly indicated the code passes validation
+            is_valid = "FAIL" not in validation.upper() and "PASS" in validation.upper()
 
             return {
                 "validation_result": {