Fix complete groupby

[SumanGouda]

PR Description

• Uses getattr(df, "obj", df).columns in _computations_complete to safely inspect column names without raising AttributeError when df is a DataFrameGroupBy object.
• Adds early validation to check if keys passed in fill_value exist in the target DataFrame.
• Filters missing column keys (col in fill_value) during dictionary construction to prevent KeyError crashes when passing partial fill_value dictionaries.
• Preserves grouping context by delaying df = df.obj unpacking until after df.expand() executes.

This PR resolves #1669.

PR Checklist

Please ensure that you have done the following:

1. PR in from a fork off your branch.
2. Add a line to CHANGELOG.md under the latest version header describing the contribution.

Relevant Reviewers

Please tag maintainers to review.

• @ericmjl

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.99%. Comparing base (901f4b3) to head (ee4007d).
⚠️ Report is 218 commits behind head on dev.

Additional details and impacted files

@@            Coverage Diff             @@
##              dev    #1700      +/-   ##
==========================================
- Coverage   87.56%   85.99%   -1.58%     
==========================================
  Files          95      125      +30     
  Lines        6819     9965    +3146     
==========================================
+ Hits         5971     8569    +2598     
- Misses        848     1396     +548     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[samukweku]

adorn.py should not be in this pr, same goes for the GitHub template.md file

[samukweku]

Kindly rebase your pr and keep out unrelated code/files