Hello,

I've been facing some issues when indexing source code using the tree-sitter-stack-graphs-python crate, particularly with failed to build stack graph errors linked to issues #426 & #403.

I'm aware these might be resolved in future updates, and I'm also considering diving into the .tsg DSL as I become more familiar with this project. Tho I believe that given the complexity of various codebases, similar errors could arise and the future, hence the idea of and having a way to safely index, ignoring potential errors.

I noticed that GitHub's code navigation still functions correctly on the very same files that are throwing errors. This leads me to believe there might be a workaround to build the stack graph despite these errors.

Is there a known method to bypass such errors during the graph building process, so the entire codebase isn't compromised by issues in single nodes? If such a feature exists or could be implemented, it would greatly enhance the tool's resilience.

Looking forward to your insights or suggestions.

PS: thank you for this amazing project and related talks / articles, it's been really enjoyable to learn about this