[decimal] Implement true_divide for BigDecimal type + fix a bug in BigUInt.multiply (#70)

This pull request includes significant changes to the BigInt and
BigDecimal arithmetic operations, introducing new functionality and
improving existing methods. The most important changes include adding
true division functionality for BigDecimal, modifying scaling methods,
and updating the handling of leading zeros in BigUInt operations.

Notably, this PR fixes a bug in `BigUInt.multiply` where the carry is
mistakenly skipped if a word of x2 is zero.

### BigDecimal Enhancements:
* Added `true_divide` function to perform division with specified
precision and rounding rules in
`src/decimojo/bigdecimal/arithmetics.mojo`.
* Updated the benchmarks to include the new division operation in
`benches/bigdecimal/bench.mojo`.
[[1]](diffhunk://#diff-96175f1000f81fe074f70a9adc4b39cd3f2dac8e85d30aa7efed9fc02b46e5ecR4)
[[2]](diffhunk://#diff-96175f1000f81fe074f70a9adc4b39cd3f2dac8e85d30aa7efed9fc02b46e5ecR16)
[[3]](diffhunk://#diff-96175f1000f81fe074f70a9adc4b39cd3f2dac8e85d30aa7efed9fc02b46e5ecR29-R35)

### BigUInt Improvements:
* Replaced `remove_trailing_zeros` with `remove_leading_empty_words` to
handle leading zeros more effectively in multiple functions across
`src/decimojo/biguint/arithmetics.mojo`.
[[1]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL180-R180)
[[2]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL275-R271)
[[3]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL383-R373)
[[4]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL964-R827)
[[5]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL1032-R895)
* Simplified the `divmod` function to use `floor_divide` and `subtract`
for quotient and remainder calculation in
`src/decimojo/biguint/arithmetics.mojo`.

### Codebase Simplification:
* Renamed `multiply_by_power_of_10` to `scale_up_by_power_of_10` for
clarity and consistency in `src/decimojo/biguint/arithmetics.mojo`.
* Removed unnecessary cases and optimized the `floor_divide` function in
`src/decimojo/biguint/arithmetics.mojo`.
[[1]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL338-L343)
[[2]](diffhunk://#diff-95a5c66e5957fff2a2a8a2710ffb11a4ad7d0bddaf3bc11d075d62ddaa8b915cL358-R348)

These changes enhance the arithmetic capabilities of BigDecimal and
BigUInt, improve code clarity, and ensure more efficient handling of
numerical operations.

Author: forfudan

benches/bigdecimal/bench.mojo

@@ -1,6 +1,7 @@
 from bench_bigdecimal_add import main as bench_add
 from bench_bigdecimal_subtract import main as bench_sub
 from bench_bigdecimal_multiply import main as bench_multiply
+from bench_bigdecimal_divide import main as bench_divide
 
 
 fn main() raises:
@@ -12,6 +13,7 @@ This is the BigInt Benchmarks
 add:         Add
 sub:         Subtract
 mul:         Multiply
+div:         Divide (true divide)
 all:         Run all benchmarks
 q:           Exit
 =========================================
@@ -24,10 +26,13 @@ q:           Exit
         bench_sub()
     elif command == "mul":
         bench_multiply()
+    elif command == "div":
+        bench_divide()
     elif command == "all":
         bench_add()
         bench_sub()
         bench_multiply()
+        bench_divide()
     elif command == "q":
         return
     else: