Don't account for crash state changes in Java stackoverflow bugs

[oliverchang]

Currently, Java stackoverflow bugs manifest as a "Security exception" crash type (example: https://oss-fuzz.com/testcase-detail/6307143094960128).

In our daily progression_task, we exclude stack-overflow bugs from having their crash states compared for reproducibility (because the stacktrace is expected to easily change across runs): https://github.com/google/clusterfuzz/blob/32a0df6e824b96df2bb2d9ad400894564461d4b6/src/clusterfuzz/_internal/bot/testcase_manager.py#L83

This is so we don't keep closing/re-opening new issues for the same crash.

We need to account for Java stack-overflows also by either:

• Changing the crash type of Java stack-overflows to "Stack-overflow"
• Somehow annotate/detect "Security exception" crashes as stack-overflow where relevant, so we can skip the crash state check.

Discovered this from https://github.com/google/oss-fuzz/issues/13177

[abohoss]

Hi @oliverchang , I would love to work on this issue but when I try to download the testcase file in the mentioned oss-fuzz issue it gives me access denied on the Oss-fuzz website

[github-actions[bot]]

This issue has not had any activity for 60 days and will be automatically closed in two weeks

[cushon]

This issue has not had any activity for 60 days and will be automatically closed in two weeks

The issue in #13177 is still happening

[github-actions[bot]]

This issue has not had any activity for 60 days and will be automatically closed in two weeks

[cushon]

This is still relevant

[github-actions[bot]]

This issue has not had any activity for 60 days and will be automatically closed in two weeks

[github-actions[bot]]

Automatically closing stale issue