core/simd/x86: Use the `none` calling convention for intrinsics

The LLVM intrinsics that live under `llvm.x86` are not actual functions,
so trying to invoke them as such using the platform's native C
calling convention causes incorrect types to be emitted in the IR.

Thanks to laytanl for assistance in testing.

core/simd/x86/adx.odin

@@ -37,7 +37,7 @@ when ODIN_ARCH == .amd64 {
 	}
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.addcarry.32")
 	llvm_addcarry_u32  :: proc(a: u8, b: u32, c: u32) -> (u8, u32) ---

core/simd/x86/fxsr.odin

@@ -21,7 +21,7 @@ when ODIN_ARCH == .amd64 {
 	}
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.fxsave")
 	fxsave    :: proc(p: rawptr) ---

core/simd/x86/pclmulqdq.odin

@@ -6,7 +6,7 @@ _mm_clmulepi64_si128 :: #force_inline proc "c" (a, b: __m128i, $IMM8: u8) -> __m
 	return pclmulqdq(a, b, u8(IMM8))
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.pclmulqdq")
 	pclmulqdq :: proc(a, round_key: __m128i, #const imm8: u8) -> __m128i ---

core/simd/x86/rdtsc.odin

@@ -11,7 +11,7 @@ __rdtscp :: #force_inline proc "c" (aux: ^u32) -> u64 {
 	return rdtscp(aux)
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.rdtsc")
 	rdtsc  :: proc() -> u64 ---

core/simd/x86/sha.odin

@@ -30,7 +30,7 @@ _mm_sha256rnds2_epu32 :: #force_inline proc "c" (a, b, k: __m128i) -> __m128i {
 	return transmute(__m128i)sha256rnds2(transmute(i32x4)a, transmute(i32x4)b, transmute(i32x4)k)
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.sha1msg1")
 	sha1msg1    :: proc(a, b: i32x4) -> i32x4 ---

core/simd/x86/sse.odin

@@ -532,7 +532,7 @@ when ODIN_ARCH == .amd64 {
 }
 
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.sse.add.ss")
 	addss       :: proc(a, b: __m128) -> __m128 ---

core/simd/x86/sse2.odin

@@ -1040,7 +1040,7 @@ when ODIN_ARCH == .amd64 {
 }
 
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name="llvm.x86.sse2.pause")
 	pause      :: proc() ---

core/simd/x86/sse3.odin

@@ -49,7 +49,7 @@ _mm_moveldup_ps :: #force_inline proc "c" (a: __m128) -> __m128 {
 	return simd.shuffle(a, a, 0, 0, 2, 2)
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name = "llvm.x86.sse3.addsub.ps")
 	addsubps :: proc(a, b: __m128) -> __m128 ---

core/simd/x86/sse41.odin

@@ -291,7 +291,7 @@ when ODIN_ARCH == .amd64 {
 }
 
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name = "llvm.x86.sse41.pblendvb")
 	pblendvb   :: proc(a, b: i8x16, mask: i8x16) -> i8x16 ---

core/simd/x86/sse42.odin

@@ -104,7 +104,7 @@ when ODIN_ARCH == .amd64 {
 	}
 }
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	// SSE 4.2 string and text comparison ops
 	@(link_name="llvm.x86.sse42.pcmpestrm128")

core/simd/x86/ssse3.odin

@@ -105,7 +105,7 @@ _mm_sign_epi32 :: #force_inline proc "c" (a, b: __m128i) -> __m128i {
 
 
 
-@(private, default_calling_convention="c")
+@(private, default_calling_convention="none")
 foreign _ {
 	@(link_name = "llvm.x86.ssse3.pabs.b.128")
 	pabsb128     :: proc(a: i8x16) -> u8x16 ---