Make Fabric and Databricks notebook run from local machine

[sonalgoyal]

Currently, we need to access the notebooks in the Databricks and Fabric environment. This makes development slow. It should be possible for us to run the notebooks through VSCode, by using their extensions.

[Srishti-8976]

Tested running Zingg Databricks notebooks from VS Code using the official Databricks extension.
VS Code ↔ Databricks integration works, but Zingg fails on Serverless compute due to JVM access restrictions (_jSparkSession).
Databricks Free/Trial workspaces only provide Serverless/SQL Warehouses, so Zingg cannot run there.
Conclusion: Zingg can be run from VS Code only on Databricks workspaces with classic/all-purpose clusters enabled.

[inshane13]

Tested running Microsoft Fabric notebooks from VS Code.

Steps performed:

• Installed Jupyter extension
• Installed Fabric Data Engineering extension
• Configured local work folder
• Signed in and selected Fabric workspace
• Downloaded notebook locally
• Selected Fabric runtime kernel
• Executed and published changes

VS Code ↔ Fabric workspace integration works (local editing, sync, publish).

Spark execution from VS Code is supported, but it runs on remote Fabric-managed compute.

There is no Fabric equivalent to Databricks Connect — full local Spark runtime execution is not supported.

Conclusion: Fabric notebooks can be developed locally in VS Code, but execution always depends on remote Fabric compute.