Fix casting floats to Decimal64 (#8363)

# Which issue does this PR close?
Closes #8362


# Rationale for this change
Fixes casting codepaths that currently fail.

# What changes are included in this PR?


# Are these changes tested?

yes

If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?

# Are there any user-facing changes?

Code should now behave as expected

---------

Co-authored-by: Andrew Lamb <andrew@nerdnetworks.org>

Author: AdamGS

arrow-cast/src/cast/decimal.rs

@@ -81,7 +81,9 @@ impl DecimalCast for i64 {
     }
 
     fn from_f64(n: f64) -> Option<Self> {
-        n.to_i64()
+        // Call implementation explicitly otherwise this resolves to `to_i64`
+        // in arrow-buffer that behaves differently.
+        num::traits::ToPrimitive::to_i64(&n)
     }
 }
 

arrow-cast/src/cast/mod.rs

@@ -3101,6 +3101,35 @@ mod tests {
         );
     }
 
+    #[test]
+    fn test_cast_floating_to_decimals() {
+        for output_type in [
+            DataType::Decimal32(9, 3),
+            DataType::Decimal64(9, 3),
+            DataType::Decimal128(9, 3),
+            DataType::Decimal256(9, 3),
+        ] {
+            let input_type = DataType::Float64;
+            assert!(can_cast_types(&input_type, &output_type));
+
+            let array = vec![Some(1.1_f64)];
+            let array = PrimitiveArray::<Float64Type>::from_iter(array);
+            let result = cast_with_options(
+                &array,
+                &output_type,
+                &CastOptions {
+                    safe: false,
+                    format_options: FormatOptions::default(),
+                },
+            );
+            assert!(
+                result.is_ok(),
+                "Failed to cast to {output_type} with: {}",
+                result.unwrap_err()
+            );
+        }
+    }
+
     #[test]
     fn test_cast_decimal128_to_decimal128_overflow() {
         let input_type = DataType::Decimal128(38, 3);