Checked narrow casts; fix runtime require in self-contained-built binaries

byte/short/int/long/char silently wrapped or passed out-of-range values
through; the JVM range-checks (RT.byteCast family). One checked-cast
helper now carries the ranges: a double range-checks ITSELF before
truncating ((byte 1.1) is 1, (byte 127.000001) throws), NaN casts to 0,
ratios and bigdecs truncate, a non-number is CCE, and the throw carries
the JVM message. float range-checks against Float/MAX_VALUE. The
unchecked-* casts now genuinely wrap and sign-fold ((unchecked-byte 200)
is -56 — the old bit-and lost the sign) with doubles saturating like
Java's conversions; unchecked-long/int are host natives. double/float of
a bigdec convert instead of crashing. The no-single-float residue stays
accepted (SPEC.md).

Also fixes #290: a binary built by the SELF-CONTAINED joltc died with
'variable var-deref is not bound' when a namespace loaded at runtime.
The in-process build compiled flat.ss against a clean copy-environment,
which orphans every top-level define in locations the binary's runtime
eval can't see. It now compiles against the default interaction
environment (defines land in the real symbol cells, same as the legacy
fresh-Chez path) and a generated prologue pre-binds each kernel name the
runtime redefines to its kernel value, so the earliest boot reads match
the legacy path's primitive references. requiring-resolve is implemented
(the issue's dynamic-require pattern), and the release workflow smokes a
runtime require in a built binary.

Cast namespaces byte/short/int/long/char now fully clean; cts baseline
5805 -> 5857 pass, 67 baselined namespaces. 7 JVM-certified corpus rows.

test/chez/corpus.edn

@@ -3551,4 +3551,11 @@
   {:suite "iref / watches" :label "add-watch returns the var; a validator gates the var root" :expected "[true :ise]" :actual "(do (def vv-y 1) [(= (var vv-y) (add-watch (var vv-y) :k (fn [a b c d] nil))) (do (set-validator! (var vv-y) pos?) (try (alter-var-root (var vv-y) -) (catch IllegalStateException _ :ise)))])"}
   {:suite "iref / watches" :label "agent watches fire on state change" :expected "[0 1]" :actual "(let [ag (agent 0) p (promise)] (add-watch ag :w (fn [k r o n] (deliver p [o n]))) (send ag inc) (deref p 2000 :timeout))"}
   {:suite "iref / meta" :label "alter-meta! works on atoms" :expected "{:x 1 :y 2}" :actual "(let [a (atom 1 :meta {:x 1})] (alter-meta! a assoc :y 2))"}
+  {:suite "casts / checked narrow" :label "doubles range-check themselves, then truncate" :expected "[1 1 3 1 \\a 0]" :actual "[(byte 1.1) (short 1.9) (int 3.7) (long 1.5) (char 97.5) (long ##NaN)]"}
+  {:suite "casts / checked narrow" :label "out-of-range values throw IllegalArgumentException" :expected "[:iae :iae :iae :iae :iae :iae :iae]" :actual "[(try (byte 128) (catch IllegalArgumentException _ :iae)) (try (byte 127.000001) (catch IllegalArgumentException _ :iae)) (try (short 40000) (catch IllegalArgumentException _ :iae)) (try (int 2147483647.000001) (catch IllegalArgumentException _ :iae)) (try (char -1) (catch IllegalArgumentException _ :iae)) (try (char 65536) (catch IllegalArgumentException _ :iae)) (try (long 1e300) (catch IllegalArgumentException _ :iae))]"}
+  {:suite "casts / checked narrow" :label "the throw carries the JVM message" :expected "\"Value out of range for byte: 128\"" :actual "(try (byte 128) (catch IllegalArgumentException e (ex-message e)))"}
+  {:suite "casts / checked narrow" :label "ratios and bigdecs truncate; non-numbers are CCE" :expected "[1 5 :cce]" :actual "[(byte 3/2) (int 5.5M) (try (byte true) (catch ClassCastException _ :cce))]"}
+  {:suite "casts / checked narrow" :label "float range-checks against Float/MAX_VALUE" :expected "[:iae 1.5]" :actual "[(try (float 1e300) (catch IllegalArgumentException _ :iae)) (float 1.5)]"}
+  {:suite "casts / unchecked wrap" :label "unchecked casts wrap and sign-fold; doubles saturate" :expected "[-56 -25536 \\A 9223372036854775807 0 0]" :actual "[(unchecked-byte 200) (unchecked-short 40000) (unchecked-char 65) (unchecked-long 1e300) (unchecked-int 4294967296) (unchecked-long ##NaN)]"}
+  {:suite "ns / requiring-resolve" :label "loads the namespace then resolves; unqualified throws" :expected "[#{1 2} :iae]" :actual "[((requiring-resolve (quote clojure.set/union)) #{1} #{2}) (try (requiring-resolve (quote unqualified)) (catch IllegalArgumentException _ :iae))]"}
 ]

test/chez/cts-known-failures.txt

@@ -6,30 +6,25 @@ clojure.core-test.add-watch 0 1
 clojure.core-test.bigint 6 0
 clojure.core-test.bit-set 1 0
 clojure.core-test.boolean-qmark 0 3
-clojure.core-test.byte 7 5
-clojure.core-test.char 1 0
 clojure.core-test.compare 1 0
 clojure.core-test.conj 1 0
 clojure.core-test.contains-qmark 3 0
 clojure.core-test.counted-qmark 1 0
 clojure.core-test.dec 1 0
 clojure.core-test.denominator 0 3
-clojure.core-test.double 0 4
 clojure.core-test.double-qmark 3 0
 clojure.core-test.empty 1 0
 clojure.core-test.eq 2 0
 clojure.core-test.eval 0 3
 clojure.core-test.even-qmark 1 0
-clojure.core-test.float 4 4
+clojure.core-test.float 1 0
 clojure.core-test.get 0 1
 clojure.core-test.ifn-qmark 2 0
 clojure.core-test.inc 1 0
-clojure.core-test.int 4 5
 clojure.core-test.int-qmark 3 0
 clojure.core-test.intern 2 0
 clojure.core-test.keys 0 4
 clojure.core-test.lazy-seq 3 0
-clojure.core-test.long 2 5
 clojure.core-test.minus 2 0
 clojure.core-test.mod 23 0
 clojure.core-test.neg-int-qmark 1 0
@@ -55,7 +50,6 @@ clojure.core-test.remove-watch 0 1
 clojure.core-test.run-bang 1 0
 clojure.core-test.select-keys 2 0
 clojure.core-test.seqable-qmark 1 0
-clojure.core-test.short 7 5
 clojure.core-test.shuffle 1 0
 clojure.core-test.some-fn 3 0
 clojure.core-test.sort-by 2 0

test/conformance/SPEC.md

@@ -132,6 +132,19 @@ arithmetic fast path (both operands Chez numbers → the raw Chez op) or force
 every `+`/`-`/`*` through an unwrapping dispatcher, de-optimizing all
 arithmetic. Same shape as the accepted BigInt-vs-Long unification.
 
+The cast RANGE contract is full parity (corpus `casts / *`): `byte`/`short`/
+`int`/`long`/`char` range-check like `RT.byteCast` — an out-of-range value is
+IllegalArgumentException "Value out of range for byte: 128". A double operand
+range-checks ITSELF before truncating (`(byte 1.1)` is `1`, `(byte 127.000001)`
+throws), NaN casts to 0, ratios and bigdecs truncate, a non-number is
+ClassCastException. `float` range-checks against Float/MAX_VALUE. The
+`unchecked-*` casts wrap and sign-fold like the JVM primitive conversions
+(`(unchecked-byte 200)` is `-56`; a double saturates instead of wrapping).
+What jolt does NOT model is a distinct single-float type: `(float x)` keeps
+the double VALUE, so a double below Float/MIN_VALUE stays nonzero and float
+rounding does not occur (the accepted no-single-float residue, baselined with
+`:integer-box-model`'s class residue).
+
 ## Number operations
 
 Binary arithmetic and comparisons follow the JVM's `Numbers.ops(x, y)` category