NMake: Document response file encoding heuristic in a comment

Since commit f3f57cc4ed (NMake: Use UTF-8 with BOM if supported by
nmake, 2021-04-22, v3.21.0-rc1~217^2) the encoding of response files is
selected based on the makefile encoding.  In principle these may be
orthogonal, but in practice it is a useful heuristic.  Call out this
heuristic in a comment, and leave a FIXME to do something better.

Source/cmMakefileTargetGenerator.cxx

@@ -16,6 +16,8 @@
 #include <cmext/algorithm>
 #include <cmext/string_view>
 
+#include "cm_codecvt.hxx"
+
 #include "cmComputeLinkInformation.h"
 #include "cmCustomCommand.h"
 #include "cmCustomCommandGenerator.h"
@@ -2046,11 +2048,17 @@ std::string cmMakefileTargetGenerator::CreateResponseFile(
   const char* name, std::string const& options,
   std::vector<std::string>& makefile_depends)
 {
+  // FIXME: Find a better way to determine the response file encoding,
+  // perhaps using tool-specific platform information variables.
+  // For now, use the makefile encoding as a heuristic.
+  codecvt::Encoding responseEncoding =
+    this->GlobalGenerator->GetMakefileEncoding();
+
   // Create the response file.
   std::string responseFileNameFull =
     cmStrCat(this->TargetBuildDirectoryFull, '/', name);
-  cmGeneratedFileStream responseStream(
-    responseFileNameFull, false, this->GlobalGenerator->GetMakefileEncoding());
+  cmGeneratedFileStream responseStream(responseFileNameFull, false,
+                                       responseEncoding);
   responseStream.SetCopyIfDifferent(true);
   responseStream << options << "\n";