feat: full native make_interval with 2.6x faster than spark and 1.1x faster than codegen dispatch

[peterxcli]

Which issue does this PR close?

Closes #5279.
Closes #5131.

Rationale for this change

Comet represented Spark CalendarIntervalType as Arrow IntervalMonthDayNano. Converting Spark's microseconds to nanoseconds reduced the valid elapsed-time range by 1,000x, while the native datafusion-spark kernel also coerced Decimal(18,6) seconds to Float64, losing microsecond precision.

Spark represents calendar intervals losslessly as separate months, days, and microseconds. Comet needs the same representation across JVM/native boundaries and exact microsecond arithmetic in the native kernel.

What changes are included in this PR?

• Represent CalendarIntervalType as a Spark-tagged Arrow struct containing months: Int32, days: Int32, and microseconds: Int64.
• Preserve that logical type through Arrow conversion, protobuf serde, FFI, JVM readers/writers, codegen input/output, Scala UDF codegen, and native execution.
• Replace the datafusion-spark make_interval wrapper with an exact Decimal(18,6) microsecond kernel with Spark-compatible NULL and ANSI/TRY overflow behavior.
• Remove the obsolete incompatibility gate, ignored regressions, and unused datafusion-spark dependency.
• Port Spark 4.2.0 make_interval boundary and arity cases with source permalinks.
• Remove the obsolete duplicate codegen-dispatch benchmark case now that make_interval is fully native.

How are these changes tested?

• cargo test --manifest-path native/Cargo.toml -p datafusion-comet-spark-expr preserves_spark_microsecond_range_and_overflow
• cargo check --manifest-path native/Cargo.toml -p datafusion-comet-spark-expr
• cargo check --manifest-path native/Cargo.toml -p datafusion-comet
• cargo fmt --manifest-path native/Cargo.toml --all -- --check
• make core
• Focused CometArrowStreamSuite CalendarInterval round-trip test
• Focused CometCodegenSuite CalendarInterval codegen test
• CometSqlFileTestSuite make_interval: 6/6 passed, 0 ignored
• CometSqlFileTestSuite calendar_interval: 1/1 passed
• Spotless, Scalastyle, and git diff --check

Benchmark

CometDatetimeExpressionBenchmark, 1,048,576 rows, Apple M4, JDK 17. The codegen-dispatch result is from parent commit 268849c0c; the full-native and Spark results are from this PR at 72997e9e8. Both Comet revisions used optimized native builds and the same query and input.

Execution path	Best time	Average time	Throughput	Per row	Relative to dispatch
Comet codegen dispatch	31 ms	33 ms	33.8 M rows/s	29.6 ns	1.00x
Comet full native	28 ms	30 ms	37.3 M rows/s	26.8 ns	1.11x
Spark	74 ms	81 ms	14.1 M rows/s	70.7 ns	0.42x

Full native is about 10% faster than codegen dispatch and 2.6x faster than Spark by best time.