Fix missing template variables in DABs bundle init

[shdzhang]

Description

The DABs installer fails when the .databrickscfg profile authenticates as a service principal. The databricks.yml.tmpl template references three variables (is_service_principal, user_name, workspace_host) that were neither declared in the template schema nor populated in the config file, causing databricks bundle init to fail with:

Error: failed to compute file content for tmp/databricks.yml.tmpl. variable "is_service_principal" not defined
./setup.sh: line 12: cd: tmp: No such file or directory
Error: unable to locate bundle root: databricks.yml not found

Additionally, the template used incorrect Go template syntax — {{workspace_host}} instead of {{.workspace_host}}.

Fix:

• Added the three missing variable declarations (is_service_principal, user_name, workspace_host) to the template schema
• Fixed Go template syntax in databricks.yml.tmpl (added . prefix to all variable references)
• Added auto-detection of the current identity in main.py using client.current_user.me(), which determines whether the caller is a service principal or a regular user and sets run_as accordingly

Type of Change

• Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

• Reproduced the original error by running ./install.sh with a service principal profile
• Verified the config file contents via debug instrumentation to confirm the missing variables
• Confirmed databricks bundle init was failing with exit code 1 and the output directory was not being created
• Applied the fix and verified successful deployment with both user and service principal authentication

Checklist

• I have reviewed the CONTRIBUTING and VERSIONING documentation.
• My code follows the code style of this project.
• I have verified, added or updated the tests to cover my changes (if applicable).
• I have verified that my changes do not introduce any breaking changes on all the supported clouds (AWS, Azure, GCP).
• I have commented my code, particularly in hard-to-understand areas.
• I have made corresponding changes to the documentation (if applicable).

Additional Notes

• The identity detection heuristic ("@" not in user_name) relies on the fact that service principals have UUID-format usernames while regular users have email addresses. This is consistent with Databricks SCIM API behavior across all clouds.
• When authenticated as a regular user, run_as uses user_name:; when authenticated as a service principal, it uses service_principal_name:.

[CLAassistant]

All committers have signed the CLA.