docs: document null-handling function arguments

[BharatDeva]

Which issue does this PR close?

Related to #1463.

Rationale for this change

Some function docstrings in python/datafusion/functions.py include examples but do not describe their arguments. This makes the generated API docs harder to scan when users are looking up function signatures.

This PR makes a small, focused pass over the related null-handling helpers.

What changes are included in this PR?

Adds Args: sections for:

• coalesce
• ifnull
• nanvl
• nvl
• nvl2

Are there any user-facing changes?

Yes, documentation-only. The generated API docs will include clearer argument descriptions for these functions. Runtime behavior is unchanged.

Validation

• uv tool run ruff@0.15.1 check --config pyproject.toml python/datafusion/functions.py
  • Result: All checks passed!
• uv tool run ruff@0.15.1 format --check --config pyproject.toml python/datafusion/functions.py
  • Result: 1 file already formatted
• git diff --check
  • Result: passed with no whitespace errors

LLM-generated code disclosure

This documentation update was prepared with assistance from OpenAI Codex and manually reviewed before submission.

[timsaucer]

I don't think this PR closes the linked issue. Also it states that these arguments are not documented but my read of each one does include documentation of what the variables are. They are not laid out in an Args: section but these are also nearly one line argument descriptions. This PR seems more pedantic than adding significant value. @BharatDeva can you expand a little on why you think the existing documentation is insufficient?

[coracuity]

Looks good, thanks so much!

[coracuity]

Ahh I didn't read the source issue, just that it sets out to do what it text here explains.

[coracuity]

@timsaucer these descriptions are very similar to what's found here https://datafusion.apache.org/user-guide/sql/scalar_functions.html#nvl

[timsaucer]

Since the original reporter is happy with it, I'm merging