fix: normalize exponent in rounding functions and support negative precision in Truncate

[amitmishra11]

Summary

This PR fixes two related bugs in decimal rounding and truncation:

1. RoundUp, RoundDown, RoundFloor, and RoundCeil returned the original decimal (with its original exponent) when the value was already exact at the requested precision, causing the result to carry a stale exponent that did not match the requested places. (Fixes Inconsistent rounding behavior in Round and RoundUp methods #390)

2. Truncate rejected negative precision values due to a precision >= 0 guard, silently returning the input unchanged. Negative precision should truncate the integer part toward zero — for example (5432).Truncate(-2) should return 5400. (Fixes Truncate should parse negative values too, or return error #406)

Changes

decimal.go

• RoundCeil: return rescaled (not d) when the value is already exact at the target precision
• RoundFloor: same
• RoundUp: same
• RoundDown: same
• Truncate: remove the precision >= 0 guard; rescale uses big.Int.Quo (truncates toward zero) so negative precision works correctly without additional adjustment; update doc comment and examples

decimal_test.go

• TestTruncateNegativePrecision: table-driven tests for Truncate with negative and positive precision, checking both value and exponent
• TestRoundingExponentNormalization: table-driven tests for all four rounding functions verifying that the returned exponent matches the requested places when the value is already exact

Testing

$ go test ./...
ok  github.com/shopspring/decimal  8.2s

Example: exponent normalization (issue #390)

Before this fix:

d, _ := decimal.NewFromString("100.0")
r := d.RoundUp(0)
r.Exponent()                     // -1  (wrong, should be 0)
r.StringFixedBank(-r.Exponent()) // "100.0"  (wrong)

After:

d, _ := decimal.NewFromString("100.0")
r := d.RoundUp(0)
r.Exponent()                     // 0
r.StringFixedBank(-r.Exponent()) // "100"

The same fix applies to RoundDown, RoundFloor, and RoundCeil.

Example: negative precision in Truncate (issue #406)

Before:

decimal.RequireFromString("5432").Truncate(-2).String() // "5432"  (unchanged, wrong)

After:

decimal.RequireFromString("5432").Truncate(-2).String()  // "5400"
decimal.RequireFromString("-5432").Truncate(-2).String() // "-5400"
decimal.RequireFromString("5499").Truncate(-2).String()  // "5400"
decimal.RequireFromString("5500").Truncate(-2).String()  // "5500"