lib/rt: rename rt_alloc → rt_malloc; rt.alloc → rt.malloc

Hare's canonical runtime allocator is rt::malloc with linker symbol
rt.malloc (ref/hare/rt/malloc.ha:27,78). ww kept the dot→underscore
Plan 9 convention (CLAUDE.md rule 4) so the linker symbol becomes
rt_malloc; the lib/rt exported function name becomes malloc; ww
callers say rt.malloc(...).

The language builtin keyword stays `alloc(T)!` — unchanged from Hare
(ref/hare/hare/lex/token.ha:21 ltok::ALLOC, parse/expr.ha:398
builtin()). The rename only touches the lowered linker symbol and the
exported function name behind it; the user-facing syntax for
heap-allocation is identical to Hare.

Surface:
- rt/alloc.s: TEXT rt_alloc → TEXT rt_malloc, labels updated
- lib/rt/malloc.ww: @symbol("rt_malloc") fn malloc(...) (was rt_alloc/alloc)
- rt/ensure.ww: local FFI decl + call site updated to malloc; `!` dropped
  on the direct FFI call (rt_malloc returns *void, not a tagged union)
- 18 .ww callers: rt.alloc(...) → rt.malloc(...)
- cstage cmd/wcc/check.c + wwstage selfhost/cmd/wcc/check.ww
  alloc-builtin suppression gate routes through ffi_resolve("malloc")
  for the lowering; the user-shadow check still keys on the BUILTIN
  KEYWORD "alloc" since that is what `alloc(...)` parses as. Adding
  "malloc" to the user-shadow check was unnecessary and was reverted
  during pre-commit review.
- cstage cmd/w6c/cgen.c: 2× ffi_resolve("alloc") → ffi_resolve("malloc")
- wwstage cgenexpr/cgenstmt: 2× ffiresolve(c, "alloc") → ffiresolve(c, "malloc")
- Test fixtures (700_e2e, 758_cgalloc_str_field, 990_selfhost, 992_w6l_ww,
  selfhost/test/tagged_ptr_ret.ww): updated inline ww sources to the new
  decl + call form

This is commit 2 of 3 in the lib/rt extraction (#38). Commit 3 closes
the OOM contract — return type becomes nullable *void and the builtin
lowering null-checks + propagates nomem.

Verified 132/132 + 995_self_rebuild byte-identity (5 wwstage tools
round-trip identical) + make clean cold rebuild.

selfhost/cmd/wcc/cgenexpr.ww

@@ -2712,14 +2712,14 @@ fn cgbin(c: *cgen, n: *node) void = {
 };
 
 // cgalloc — `alloc(value)` builtin lowering. Allocate sizeof(value)
-// bytes via rt_alloc, then write the value's bytes into the new
+// bytes via rt_malloc, then write the value's bytes into the new
 // region. For an N_STRUCTLIT arg, allocate the struct's totsize and
 // emit per-field stores at each field's offset. For a scalar/ptr,
 // allocate 8 bytes and store one word. Mirrors cmd/w6c/cgen.c's
 // alloc-special branch in N_CALL.
 //
 // Task #30: result is the graduated `(*T | nomem)` tagged-pointer
-// pair (AX=tag, DX=ptr). rt_alloc now returns 0 on OOM
+// pair (AX=tag, DX=ptr). rt_malloc now returns 0 on OOM
 // (rt/alloc.s); branch on AX to emit the nomem variant (tag=1,
 // DX=0) or the success variant (tag=0, DX=ptr) after the
 // value-init stores complete. Callers wrap with `!` / `?` to
@@ -2745,7 +2745,7 @@ fn cgalloc(c: *cgen, n: *node) void = {
 	emitint(sz: i64);
 	emitline(", DI\n");
 	emitline("\tCALL\t");
-	emitline(ffiresolve(c, "alloc"));
+	emitline(ffiresolve(c, "malloc"));
 	emitline("(SB)\n");
 	emitline("\tCMPQ\t$0, AX\n");
 	emitline("\tJNE\t"); emitline(okl); emitline("\n");
@@ -2972,7 +2972,7 @@ fn cgcall(c: *cgen, n: *node) void = {
 				};
 			};
 			// `alloc(value)` builtin: heap-init a fresh *T with the
-			// value's bytes. For struct literals, lower to rt_alloc
+			// value's bytes. For struct literals, lower to rt_malloc
 			// + per-field stores. Mirrors cmd/w6c/cgen.c's N_CALL
 			// alloc path.
 			//