feat: Add /tesseract to PYTHONPATH, similar to native execution

docs/content/creating-tesseracts/advanced.md

@@ -150,6 +150,10 @@ After that is done, you will be able to use the `tesseract-runtime` command in y
 This is the exact same command that is launched inside Tesseract containers to run their
 various endpoints, and its syntax mirrors the one of `tesseract run`.
 
+```{note}
+When running without containerization, the directory containing `tesseract_api.py` is automatically added to `PYTHONPATH`, allowing you to import additional Python modules from that directory. This behavior matches what happens inside Tesseract containers, where the path containing `tesseract_api.py` is also added to `PYTHONPATH`.
+```
+
 For instance, to call the `apply` function, rather than first building a `helloworld` image and running this command:
 
 ```bash

docs/content/examples/building-blocks/localdependency.md

@@ -47,6 +47,10 @@ We can then specify it as a local dependency in `tesseract_requirements.txt` by
 ./cowsay-6.1-py3-none-any.whl
 ```
 
+```{note}
+If you have additional Python modules in the same directory as `tesseract_api.py`, you can import them directly without any special configuration. The directory containing `tesseract_api.py` is automatically added to `PYTHONPATH` in the container environment.
+```
+
 Finally, let's build the Tesseract, and verify it works
 ```bash
 $ tesseract build .

tesseract_core/sdk/templates/Dockerfile.base

@@ -83,6 +83,7 @@ COPY "{{ tesseract_source_directory }}/tesseract_api.py" ${TESSERACT_API_PATH}
 RUN chmod 444 ${TESSERACT_API_PATH}
 
 ENV PATH="/python-env/bin:$PATH"
+ENV PYTHONPATH="/tesseract:$PYTHONPATH"
 
 {% if config.build_config.package_data %}
 # Copy package data to image