Fix declaration of scalar sincos() function. (#996)

The function was accidentally defined with a generic `int` parameter copy-pasted  from the vector definition, but that made the scalar version impossible to call with inferred generic arguments, because there wasn't a way to infer `N` when it isn't used in the parameter list.

Includes a simple test case to confirm that the front-end no longer chokes on calls to scalar `sincos()`.

1 files changed, 19 insertions, 0 deletions

diff --git a/tests/front-end/sincos.slang b/tests/front-end/sincos.slang
new file mode 100644
index 000000000..b7e7beaad
--- /dev/null
+++ b/tests/front-end/sincos.slang
@@ -0,0 +1,19 @@
+// sincos.slang
+//TEST:SIMPLE:
+
+// Just confirming that calls to the `sincos()` built
+// in work in the front-end.
+
+float test( float4x4 m, float4 v, float a )
+{
+    float4x4 mc, ms;
+    sincos(m, mc, ms);
+
+    float4 vc, vs;
+    sincos(v, vc, vs);
+
+    float c, s;
+    sincos(a, c, s);
+	
+	return c + s + vc[0] + vs[0] + mc[0][0] + ms[0][0];
+}