rt: bump-over-mmap allocator (rt_segmalloc + rt_free no-op)

rt_malloc was a bare mmap-per-call wrapper: every allocation, even a
32-72B AST/symbol node, consumed a page-rounded 4KB region and was
never freed. w6a/w6c emit ~4 such nodes per .s line, so assembling a
145K-line file burned ~625K pages (~2.5GB); test 995's 5 concurrent
self-rebuilds then OOM'd. The defect is linear and str-independent --
the str->24B codegen merely enlarged .s files past the cliff.

Replace it with the no-free SUBSET of Hare's allocator
(ref/hare/rt/malloc.ha): 2MiB chunk-bump (CHUNKSZ malloc.ha:24, ALIGN
malloc.ha:14), oversized (>CHUNKSZ) requests direct-mmap'd. The
bin/freelist/META machinery exists only to support free, which ww does
not have, so it is omitted. Policy lives in rt/malloc.ww; the raw mmap
primitive stays in rt/alloc.s as rt_segmalloc -- Hare's malloc/segmalloc
split.

rt_free becomes a documented no-op (os.free re-exports it for the public
API, so the symbol must stay); rt/ensure.ww drops its now-impossible
reclaim. Zero-init is preserved: the bump never reuses memory, so every
byte is fresh MAP_ANONYMOUS-zeroed.

w6a_ww on a 145K-line .s: 2165MB -> 28MB (~glibc parity, C w6a 24MB).
.o output byte-identical; both stages emit identical asm. make test
134/134.

rt/ensure.ww

@@ -20,7 +20,6 @@
 // bare symbol name so the linker resolves it.
 
 @symbol("rt_malloc") fn malloc(n: u64) *void;
-@symbol("rt_free")  fn free(p: *void, n: u64) void;
 
 // Mirrors ww's []T header layout: 24 bytes with 8-byte slots.
 // ww's source uses i32 for len/cap but the compiler stores them in
@@ -46,9 +45,8 @@ export fn rt_ensure(s: *slice, membsz: u64) void = {
 		np[i] = s.ptr[i];
 		i += 1u64;
 	};
-	if (s.cap > 0i64) {
-		free(s.ptr: *void, (s.cap: u64) * membsz);
-	};
+	// No free: the bump allocator (rt/malloc.ww) can't reclaim a
+	// mid-chunk region; the old buffer leaks until process exit. (#8)
 	s.ptr = np;
 	s.cap = nc;
 };