ww+wcc: graduate selfhost TK_* defs to tkind enum

`type tkind = enum i32 { TK_NONE = 0, TK_EOF = 1, ... TK_LAST = 86 }`
replaces the 87-line `def TK_*: i32 = N` cluster in lib/ww/lex/tok.ww.
Numeric values explicit so 990_selfhost's byte-diff against the C-side
`Tkind` enum still passes.

All ~270 reference sites in lib/ww and selfhost/cmd/{wcc,wwdump}
sed-renamed `TK_X` → `tkind.TK_X`. Struct fields (`tok.kind`,
`parser.curkind`) intentionally kept as `i32` — making them `tkind`
shifted some byte-positions in the cgen output and broke 990/993/995
byte-identity probes without an obvious win.

To make the rename non-cascading on every signature, type_assignable
and unify_arith in cmd/wcc/check+type relax to allow enum ↔ int
mixing when storage matches (a `tkind` value flows into an `i32`
slot and vice versa, no explicit cast). This deviates from Hare's
strict enum semantics; doc'd as an explicit pragmatic relaxation
for the compiler's internal enum-shaped kinds. External user code
can still get the type-safety benefit if they declare their
parameters with the enum type.

combined.ww files regenerated by ww build.

selfhost/cmd/wcc/cgenstmt.ww

@@ -403,7 +403,7 @@ fn cglet(c: *cgen, n: *node) void = {
 		// For each field in the lit, evaluate its value and store at
 		// the field's offset within the slot. Field-name → offset
 		// from the struct registry. When the literal carries
-		// op == TK_ELLIPSIS (autofill marker from the parser), the
+		// op == tkind.TK_ELLIPSIS (autofill marker from the parser), the
 		// entire slot is zero-filled first so unmentioned fields
 		// read as 0.
 		if (rhs.kind == N_STRUCTLIT) {
@@ -416,7 +416,7 @@ fn cglet(c: *cgen, n: *node) void = {
 			};
 			let si: *structinfo = structlookup(c, sname);
 			if (si != nil) {
-				if (rhs.op == TK_ELLIPSIS) {
+				if (rhs.op == tkind.TK_ELLIPSIS) {
 					let total: i32 = si.totsize;
 					emitline("\tXORQ\tAX, AX\n");
 					let zi: i32 = 0;