Add support for InvalidFunctionArg lgamma() (#5845)

Reference: https://en.cppreference.com/w/cpp/numeric/math/lgamma

Author: orbitcowboy

cfg/std.cfg

@@ -3035,6 +3035,7 @@ The obsolete function 'gets' is called. With 'gets' you'll get a buffer overrun
       <not-uninit/>
     </arg>
   </function>
+  <!-- https://cplusplus.com/reference/cmath/lgamma/ -->
   <!-- double lgamma(double x); -->
   <function name="lgamma,std::lgamma">
     <use-retval/>
@@ -3044,8 +3045,11 @@ The obsolete function 'gets' is called. With 'gets' you'll get a buffer overrun
     <leak-ignore/>
     <arg nr="1" direction="in">
       <not-uninit/>
+      <!-- If x is zero or a negative integer for which the function is asymptotic, it may cause a pole error (depending on implementation).-->
+      <valid>!0.0:</valid>
     </arg>
   </function>
+  <!-- https://cplusplus.com/reference/cmath/lgamma/ -->
   <!-- float lgammaf(float x); -->
   <function name="lgammaf,std::lgammaf">
     <use-retval/>
@@ -3055,8 +3059,11 @@ The obsolete function 'gets' is called. With 'gets' you'll get a buffer overrun
     <leak-ignore/>
     <arg nr="1" direction="in">
       <not-uninit/>
+      <!-- If x is zero or a negative integer for which the function is asymptotic, it may cause a pole error (depending on implementation).-->
+      <valid>!0.0:</valid>
     </arg>
   </function>
+  <!-- https://cplusplus.com/reference/cmath/lgamma/ -->
   <!-- long double lgammal(long double x); -->
   <function name="lgammal,std::lgammal">
     <use-retval/>
@@ -3066,6 +3073,8 @@ The obsolete function 'gets' is called. With 'gets' you'll get a buffer overrun
     <leak-ignore/>
     <arg nr="1" direction="in">
       <not-uninit/>
+      <!-- If x is zero or a negative integer for which the function is asymptotic, it may cause a pole error (depending on implementation).-->
+      <valid>!0.0:</valid>
     </arg>
   </function>
   <!-- double rint(double x); -->

test/cfg/std.cpp

@@ -2382,6 +2382,30 @@ void uninitvar_ldexp(void)
     (void)std::ldexp(ldc,e3);
 }
 
+void invalidFunctionArg_lgamma(float f, double d, long double ld)
+{
+    (void)lgamma(d);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgamma(-0.1);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgamma(0.0);
+    (void)lgamma(0.1);
+
+    (void)lgammaf(f);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgammaf(-0.1f);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgammaf(0.0f);
+    (void)lgammaf(0.1f);
+
+    (void)lgammal(ld);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgammal(-0.1L);
+    // cppcheck-suppress invalidFunctionArg
+    (void)lgammal(0.0L);
+    (void)lgammal(0.1L);
+}
+
 void uninitvar_lgamma(void)
 {
     float f;