feat(asm): encode the arm64 Q-width FP load and store

FMOVQ routes through the MOV load/store machinery in the plain, post-index,
pre-index and static-symbol forms.  The Q width carries its size in the opc
field, so the store spelling is opc=10 and the access scales by sixteen;
both come from helpers now instead of the size exponent.  The static-symbol
form takes the toolchain's twelve-byte ADRP + ADD + access fallback with the
R_ADDRARM64 pair.  The register-to-register and immediate forms stay
rejected, matching the toolchain's own table.

Assisted-by: GLM 5.3 Flash

asm/arm64_encode_test.go

@@ -679,6 +679,42 @@ func TestArm64AcquireRelease(t *testing.T) {
 	}
 }
 
+// TestArm64QMove pins the Q-width FP move against the toolchain words the
+// corpus records: the plain, writeback and static-symbol load/store forms.
+// FMOVQ carries no register-to-register or immediate form, and both are
+// rejected the way the toolchain rejects them.
+func TestArm64QMove(t *testing.T) {
+	got := arm64Words(t, "\tFMOVQ.P F13, 11(R10)\n\tFMOVQ.W F15, 11(R20)\n\tFMOVQ.P 11(R10), F13\n\tFMOVQ.W 11(R20), F15\n"+
+		"\tFMOVQ F0, 32(R5)\n\tFMOVQ F10, 65520(R10)\n\tFMOVQ 32(R5), F2\n")
+	want := []uint32{
+		0x3c80b54d, // FMOVQ.P F13, 11(R10)
+		0x3c80be8f, // FMOVQ.W F15, 11(R20)
+		0x3cc0b54d, // FMOVQ.P 11(R10), F13
+		0x3cc0be8f, // FMOVQ.W 11(R20), F15
+		0x3d8008a0, // FMOVQ F0, 32(R5)
+		0x3dbffd4a, // FMOVQ F10, 65520(R10)
+		0x3dc008a2, // FMOVQ 32(R5), F2
+		0xd65f03c0,
+	}
+	if len(got) != len(want) {
+		t.Fatalf("word count = %d, want %d", len(got), len(want))
+	}
+	for i := range want {
+		if got[i] != want[i] {
+			t.Errorf("word %d = %08x, want %08x", i, got[i], want[i])
+		}
+	}
+	for _, src := range []string{"\tFMOVQ F1, F2\n", "\tFMOVQ $1, R2\n", "\tFMOVQ $0, 8(R3)\n"} {
+		f, errs := parser.Parse("test_arm64.s", "#include \"textflag.h\"\n\nTEXT ·f(SB), NOSPLIT, $0-0\n"+src+"\tRET\n")
+		if len(errs) > 0 {
+			t.Fatalf("parse %q: %v", src, errs)
+		}
+		if _, err := AssembleFileARM64(f); err == nil {
+			t.Errorf("%q: assembled, want rejection", strings.TrimSpace(src))
+		}
+	}
+}
+
 // TestArm64BTI pins the landing-pad family against the toolchain words:
 // only the uppercase C/J/JC spellings assemble, and bare BTI is a
 // diagnostic, never a panic.