Binary adapter for format-independent interval arithmetic

The generic finite interval API uses exact rational decoding and directed binary rounding. It returns none if an endpoint is exceptional or no finite enclosure is produced. The richer Binary.Interval operations remain available when an IEEE whole-range infinity fallback is desired instead.

def FloatLib.Floats.ExecFloat.Binary.intervalRounding {format : Formats.BinaryInterchange.FloatFormat} {plan : Formats.BinaryInterchange.Configured.StoragePlan format} {code : Type} [ModelCodec plan (Formats.BinaryInterchange.Model format) code] : Numerics.OutwardRounding (ExecFloat (Formats.BinaryInterchange.Configured.Family format code plan)) ℚ

Checked finite outward rounding for the configured binary carrier, including finite encodings.

Directed model rounding supplies candidates; exact rational comparisons certify the accepted enclosure and reject inappropriate saturation at the range boundary.