[BUGFIX] Fix crash rendering results with unexpected_index_column_names and no domain column

[Bharath-970]

• Description of PR changes above includes a link to an existing GitHub issue
• PR title is prefixed with [BUGFIX]
• Code is linted - ran ruff format + ruff check on the changed files (clean)
• Appropriate tests and docs have been updated

Closes #11933.

Problem

When result_format=COMPLETE and unexpected_index_column_names is specified, but no separate domain/value column is tracked alongside the ID/PK columns, rendering the validation result for Data Docs crashes:

ValueError: No group keys passed!

Root cause

_convert_unexpected_indices_to_df (great_expectations/render/util.py) computes domain_column_name_list as whatever keys of an unexpected_index_list entry are not in unexpected_index_column_names. When every key in the dict is an ID/PK column -- i.e. there's no separate domain column present -- that list ends up empty, and pandas.DataFrame.groupby([]) raises.

Fix

Branch on whether there's actually a domain column to group by:

if domain_column_name_list:
    all_unexpected_indices = unexpected_index_df.groupby(domain_column_name_list).agg(_agg_func)
else:
    # No domain column to group by -- each row already stands on its own;
    # wrap its values in a single-item list to match the shape
    # .groupby().agg(_agg_func) would otherwise produce.
    all_unexpected_indices = unexpected_index_df.map(lambda x: [x])

With no domain column, every unexpected row is already its own group of one -- there's nothing to merge across rows, so this skips straight to the equivalent output shape instead of crashing. The existing grouped path (domain column present) is untouched.

Test

Added test_convert_unexpected_indices_to_df_no_domain_column in tests/render/test_util.py, directly exercising the previously-crashing input shape (matches the existing test style in that file).

Confirmed as a true regression guard: reverting the fix reproduces the exact original ValueError: No group keys passed!; restoring it passes.

Verification

• Reproduced the original crash directly against _convert_unexpected_indices_to_df
• Confirmed the fix resolves it with correct output, and the existing grouped (domain-column-present) path is unaffected
• tests/render/test_util.py: 22 passed
• Broader tests/render/: 168 passed, 2 xfailed (pre-existing, unrelated)
• ruff check / ruff format --check: clean

[netlify]

👷 Deploy request for niobium-lead-7998 pending review.

Visit the deploys page to approve it

Name	Link
🔨 Latest commit	a91b718

[Bharath-970]

Opened against the wrong repo by mistake — this PR targets a fork (fivetran/great_expectations), not upstream. Closing and reopening at great-expectations/great_expectations.

[Bharath-970]

Apologies for the churn — I incorrectly assumed fivetran/great_expectations was a fork rather than the actual current home of this project (renamed/transferred from great-expectations/great_expectations, with GitHub auto-redirecting). Reopening this one; closed the accidental duplicate (#11948).