Description of the problem

We had several issues on the logs :

1. first when we had a raise ValueError(...) for example in the processChunk then on the log we only had the error and not the full log
2. also there was sometimes logs being overwritten and thus missing in the log file
3. there was additional log files (in addition to 0.log) in parallelized nodes with 1 chunk
4. finally sometimes we had this kind of thing appearing in the log :

Why we had the issues and how this PR addresses them

• for (1) the reason is that the log file was opened in write mode and therefore changing it to append on executeChunkCommandLine fixes it
• for (2) the main reason is that we were systematically clearing the file when we call chunk.logManager.start
• for (3) I made sure we never use log. Even on cases where we have a single chunk I think it's simply better to use 0.log that way we have a better naming consistency anyways.
• for (4) I don't have a defenitive explanation but the logs were written in binary mode and for some reason the QML part was adding some unknown characters while writing thee log on the log tab. Changing the file opening mode to append is what fixed the issue