LeetCode resource exhaustion ambiguity across language runtimes

LeetCode utilizes containerized environments to execute code with strict constraints on execution time and memory. When a solution fails with a 'Time Limit Exceeded' or 'Memory Limit Exceeded' error, the platform does not distinguish between an infinite loop and algorithmic complexity that merely exceeds the defined threshold.

There is currently no documented mechanism to access heap profiles or identify the specific line of code where peak allocation triggers a crash. Furthermore, memory overhead varies significantly between runtimes like Python and C++, making it difficult to determine if a failure is due to logic-based leaks or environment-specific runtime overhead.

Is there a method to determine the specific memory allocation threshold for a given runtime? How can users differentiate between a memory leak and a high-complexity algorithm when no granular telemetry is provided?