Hi,
I have been working with data visualisations in Databricks the last week and came across a very peculiar behaviour where it seems like Databricks is actively suppressing errors to the user.
An example:
```
import matplotlib.pyplot as plt
x = [1, 2, 3]
y = [1, 2, 3]
plt.plot(x, y)
x.length
```
produces only the plot as output and I get no information that the code after the plot command has failed. I have attached an image of the output. If I put the `x.length` in its own cell, it will raise the Attribute Error as expected.
Expected behaviour would be that the plot and the Attribute Error is shown. I have attached an image of this when using just ordinary Jupyter Lab.
When reading about Databricks notebooks and errors, I cannot find any information in the documentation that this is expected behaviour. I just searched for "error" but found no indication that Databricks wants to make users aware of this?
- Have I just configured something incorrectly in my user to create this very unwanted behaviour?
- If not, why on earth would Databricks do this? Why is this a feature?
I am working in DBR 7.3 LTS ML with Spark 3.0.1 and Scala 2.12 (my company is upgrading the version to DBR 10.4 LTS at the start of 2023).
Thank you.
