ffi M5.A.next.1d.A: pack impl matching — lock in concrete-only miss

Step 1d lock-in test pinning today's matching behaviour.
`registerParamImpl` records every impl in `param_impl_map` keyed
by `"Proto\x00<arg_mangled>\x00<src_mangled>"`. For a pack impl
`Into(Block) for Closure(..$args) -> $R` the key contains the
pack-shaped closure's mangle (interns with `pack_start = Some(0)`
after 1c.B). At the `xx cl : *Block` site the lookup mangles the
concrete `Closure(s32, bool) -> bool` source and finds nothing —
the existing focused diagnostic fires:

  no `Into(Block) for cl_s32_bool__bool` impl — add a per-signature
  `__block_invoke_<sig>` trampoline + Into impl alongside the
  existing ones in modules/std/objc_block.sx, or declare it in
  your own code

The pack impl is reachable in the file but never considered.

Next commit (1d.B):
- New `param_impl_pack_map` keyed by `"Proto\x00<arg_mangled>"`
  (no src) — populated by `registerParamImpl` when the source
  is pack-shaped.
- `tryUserConversion` walks the pack map on concrete-key miss.
  Pack shape matches when the impl's fixed prefix equals the
  source's matching prefix; the remainder binds to `$args` and
  the source's return type binds to `$R`. Concrete impls win
  over pack impls (specificity).
- `resolveTypeWithBindings` learns the closure_type_expr path
  so the impl body's `self: Closure(..$args) -> $R` substitutes
  to the concrete source closure during monomorphisation.

The `Closure(s32, bool) -> bool` shape is not covered by stdlib
or 96-block-multi-arg's hand-rolled impls, so the pack impl is
the only candidate post-1d.B.

193/193 example tests + `zig build test` green.

examples/155-pack-impl-match.sx

@@ -0,0 +1,46 @@
+// Variadic heterogeneous type packs — `..$args` — impl matching
+// lock-in.
+//
+// Step 1d slice for the pack feature (see
+// ~/.claude/plans/lets-see-options-for-merry-dijkstra.md). Pins
+// today's concrete-only impl-matching behaviour: a user-declared
+// `impl Into(Block) for Closure(..$args) -> $R` does NOT match
+// any concrete closure source type. The xx cast site hits the
+// existing "no `Into(Block) for <src>` impl" diagnostic even
+// though the pack impl is reachable.
+//
+// Next commit (1d.B) teaches `tryUserConversion` /
+// `registerParamImpl` to walk a second `param_impl_pack_map`
+// when the concrete-key miss happens. Pack impls bind their
+// `$args` and `$R` to the concrete source closure's tail types
+// and return; monomorphisation proceeds against those bindings.
+//
+// The `Closure(s32, bool) -> bool` shape is not covered by
+// stdlib's hand-rolled impls (only `Closure() -> void` and
+// `Closure(bool) -> void`) and not covered by 96-block-multi-arg's
+// `Closure(s32, *void) -> void` impl, so the pack impl is the
+// only candidate.
+
+#import "modules/std.sx";
+#import "modules/std/objc_block.sx";
+
+impl Into(Block) for Closure(..$args) -> $R {
+    convert :: (self: Closure(..$args) -> $R) -> Block {
+        .{
+            isa = @_NSConcreteStackBlock,
+            flags = 0,
+            reserved = 0,
+            invoke = null,
+            descriptor = xx @__sx_block_descriptor,
+            sx_env = self.env,
+            sx_fn = self.fn_ptr,
+        };
+    }
+}
+
+main :: () -> s32 {
+    cl := (a: s32, b: bool) => true;
+    b : *Block = xx cl;
+    print("pack impl match ok\n");
+    return 0;
+}