Patch dbutils.notebook.entry_point... to return current local notebook path from env var

[kartikgupta-db]

Changes

• Users are using dbutils.notebook.entry_point.getDbutils().notebook().getContext().notebookPath().get() to get the current notebook path in notebooks (referring https://stackoverflow.com/questions/53523560/databricks-how-do-i-get-path-of-current-notebook).
• This does not work from our current dbutils proxy. This PR patches this to read from a DATABRICKS_SOURCE_FILE env var when using local dbutils.

Proposal

• Allow users to add their own patches to make is easier for users using dbutils from sdk to workaround such issues in the future.

Tests

• Unit tests

• make test run locally
• make fmt applied
• relevant integration tests applied

[github-actions]

This PR breaks backwards compatibility for databrickslabs/ucx downstream. See build logs for more details.

_{Running from downstreams #81}

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 87.75510% with 6 lines in your changes are missing coverage. Please review.

Project coverage is 57.74%. Comparing base (a1055d1) to head (f914957).
Report is 42 commits behind head on main.

Files	Patch %	Lines
databricks/sdk/dbutils.py	87.75%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #618      +/-   ##
==========================================
- Coverage   57.89%   57.74%   -0.16%     
==========================================
  Files          44       47       +3     
  Lines       27956    32418    +4462     
==========================================
+ Hits        16185    18719    +2534     
- Misses      11771    13699    +1928     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[mgyucht]

Few nits but otherwise looks good.

[mgyucht]

Is this necessary? We just return result.result every time. Might as well return the result itself?

[kartikgupta-db]

This is to provide a type that wraps a None result from the override. Some of the functions can then safely return None, if the override is not found or is not able to run.

[github-actions]

This PR breaks backwards compatibility for databrickslabs/blueprint downstream. See build logs for more details.

_{Running from downstreams #81}