HDRenderPipeline: Add comment to SSS and introduce min fp16 value

Assets/ScriptableRenderPipeline/HDRenderPipeline/Lighting/Deferred.shader

                 outputs.specularLighting = float4(specularLighting, 1.0);
                 outputs.diffuseLighting  = diffuseLighting;
                 #if defined(LIGHTLOOP_TILE_INDIRECT) || defined(LIGHTLOOP_TILE_ALL)
-                    // Force non-0 indirect lighting to avoid SSS artifacts.
-                    outputs.diffuseLighting.r = max(outputs.diffuseLighting.r, 0.000001);
+                    // We SSSSS is enabled with use split lighting.
+                    // SSSSS algorithm need to know which pixels contribute to SSS and which doesn't. We could use the stencil for that but it mean that it will increase the cost of SSSSS
+                    // A simpler solution is to add a slight contribution here that isn't visible (here we chose fp16 min (which is also fp11 and fp10 min).
+                    // The SSSSS algorithm will check if diffuse lighting is black and discard the pixel if it is the case
+                    outputs.diffuseLighting.r = max(outputs.diffuseLighting.r, HFLT_MIN);
                 #endif
             #else
                 outputs.combinedLighting = float4(diffuseLighting + specularLighting, 1.0);

Assets/ScriptableRenderPipeline/HDRenderPipeline/Material/Lit/Resources/CombineSubsurfaceScattering.shader

                     /* Our blur is energy-preserving, so 'centerWeight' should be set to 0.  */ \
                     /* We do not terminate the loop since we want to gather the contribution */ \
                     /* of the remaining samples (e.g. in case of hair covering skin).        */ \
+                    /* Note: See comment in the output of deferred.shader                    */ \
                     /*************************************************************************/ \
                 }                                                                               \
             }

Assets/ScriptableRenderPipeline/ShaderLibrary/Common.hlsl

 #define FLT_MIN     1.175494351e-38 // Minimum representable positive floating-point number
 #define FLT_MAX     3.402823466e+38 // Maximum representable floating-point number
 
+#define HFLT_MIN    0.00006103515625 // 2^14  it is the same for 10, 11 and 16bit float. ref: https://www.khronos.org/opengl/wiki/Small_Float_Formats
+
 float DegToRad(float deg)
 {
     return deg * PI / 180.0;