[AURON #2485] Support last_day function natively

[Sigma-Ma]

Which issue does this PR close?

Closes #2485

Rationale for this change

Spark last_day expressions cannot currently be converted to native expressions, causing affected plans to fall back to Spark execution.

What changes are included in this PR?

• Convert Spark LastDay expressions to the Spark_LastDay extension function.
• Implement Spark_LastDay using the existing Date32 conversion and month-length helpers.
• Add Rust unit tests and Spark regression coverage for regular months, leap years, dates before the Unix epoch, and null values.

Are there any user-facing changes?

No. The Spark SQL syntax and result semantics remain unchanged. Eligible last_day expressions can now run natively.

How was this patch tested?

cargo test -p datafusion-ext-functions \
  spark_dates::tests::test_spark_last_day -- --exact

./dev/reformat --check

./auron-build.sh \
  --pre \
  --clean true \
  --sparkver 3.5 \
  --scalaver 2.12 \
  --skiptests false \
  -DwildcardSuites=org.apache.auron.AuronFunctionSuite

Was this patch authored or co-authored using generative AI tooling?

• Yes
• No

Generated-by: OpenAI Codex (GPT-5)

ASF guidance: https://www.apache.org/legal/generative-tooling.html

[slfan1989]

Thanks for adding native support for last_day. I reviewed the expression conversion, function registration, native implementation, and test coverage. The implementation correctly handles month boundaries, leap years, pre-epoch dates, and null values. All checks are passing. LGTM!

[slfan1989]

@Sigma-Ma Thanks for the contribution! Merged into master.