perf: decode rewrite, encode zeros optimization, iai-callgrind benchmarks (0.3.1)

Decode rewritten with shrinking-slice pattern (split_first/split_at)
instead of index arithmetic. Decode zeros 14% faster, decode nonzero
flips from 5% behind corncobs to 2% ahead, decode mixed closes to tied.

Encode zeros fast path moved before split_at — now tied with cobs
(was 6% behind). Code size reduced to 821B (was 869B).

Replaced criterion with iai-callgrind for deterministic instruction-count
benchmarks via Valgrind. Zero run-to-run variance. Expanded to 162
benchmarks across 9 sizes.

ucobs now wins or ties 16 of 18 benchmarks vs corncobs and cobs.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "ucobs"
-version = "0.3.0"
+version = "0.3.1"
 edition = "2021"
 rust-version = "1.93"
 license = "MIT"
@@ -26,7 +26,7 @@ all-features = true
 [dev-dependencies]
 cobs = "0.5"
 corncobs = "0.1"
-criterion = { version = "0.8", features = ["html_reports"] }
+iai-callgrind = "0.16"
 proptest = "1"
 rand = "0.10"