patched overflow in setBit (#623)

Since `bitValue` is an `int`, the expression `bitValue << bitIndex` can overflow when `bitIndex` equals/exceeds 32. This is solved by casting `bitValue` into a `qindex` before shifting.

(Tyson updated authorlists)

Author: JPRichings

AUTHORS.txt

@@ -44,6 +44,8 @@ Dr Ian Bush [consultant]
     HPC
 
 External contributors:
+James Richings
+    patched overflow in bitwise.hpp logic
 Luc Jaulmes
     patched v4's install process using CMake
 Jakub Adamski

README.md

@@ -257,6 +257,7 @@ See the [docs](docs/README.md) for enabling acceleration and running the unit te
 
 In addition to QuEST's [authors](AUTHORS.txt), we sincerely thank the following external contributors to QuEST.
 
+- [James Richings](https://github.com/JPRichings) for patching a v4 overflow bug.
 - [Luc Jaulmes](https://github.com/lucjaulmes) for patching v4's CMake installation.
 - [Jakub Adamski](https://github.com/jjacobx) for optimising distributed communication of max-size messages.
 - [Bruno Villasenor Alvarez](https://github.com/bvillasen) of [AMD](https://www.amd.com/en.html) for porting the v3 GPU backend to [HIP](https://github.com/ROCm-Developer-Tools/HIP), for compatibility with AMD GPUs.

quest/src/core/bitwise.hpp

@@ -4,6 +4,7 @@
  * 
  * @author Tyson Jones
  * @author Erich Essmann (improved OS agnosticism)
+ * @author James Richings (patched setBit)
  */
 
 #ifndef BITWISE_HPP
@@ -106,7 +107,8 @@ INLINE qindex insertBit(qindex number, int bitIndex, int bitValue) {
 
 INLINE qindex setBit(qindex number, int bitIndex, int bitValue) {
 
-    qindex bitInPlace = bitValue   << bitIndex;
+    // beware that shifting the raw int would overflow (#623)
+    qindex bitInPlace = ((qindex) bitValue) << bitIndex;
     qindex oneInPlace = QINDEX_ONE << bitIndex;
     return (number & ~oneInPlace) | bitInPlace;
 }