Update build instructions and floating-point documentation

Describes the build instructions for CMake better.

A lot of the build instructions were for float, so that section got reworked too.

The float concept stuff was moved into the Numbers document out of the header files because it is for both encoding and decoding.


* Update the build instructions

* Update the build instructions

* Lots of text updating

* Float and build documentation update

* Last merge went wrong

---------

Co-authored-by: Laurence Lundblade <lgl@securitytheory.com>

Author: laurencelundblade

CMakeLists.txt

@@ -41,6 +41,21 @@ option(QCBOR_DISABLE_INDEFINITE_LENGTH_ARRAYS  "Disable encoding and decoding of
 option(QCBOR_DISABLE_DECODE_CONFORMANCE        "Exclude all decode conformance checking features" OFF)
 option(USEFULBUF_DISABLE_STREAMING             "Exclude features for streaming encoding" OFF)
 
+# This is backwards compatibility for float-related cmake options -- decided
+# to just make the cmake option the same as the #define
+if(DEFINED QCBOR_OPT_DISABLE_PREFERRED_FLOAT)
+    set(QCBOR_DISABLE_PREFERRED_FLOAT ${QCBOR_OPT_DISABLE_PREFERRED_FLOAT} CACHE BOOL "xx" FORCE)
+endif()
+
+if(DEFINED QCBOR_OPT_DISABLE_ALL_FLOAT)
+    set(USEFULBUF_DISABLE_ALL_FLOAT ${QCBOR_OPT_DISABLE_ALL_FLOAT} CACHE BOOL "xx" FORCE)
+endif()
+
+if(DEFINED QCBOR_OPT_DISABLE_FLOAT_HW_USE)
+    set(QCBOR_DISABLE_FLOAT_HW_USE ${QCBOR_OPT_DISABLE_FLOAT_HW_USE} CACHE BOOL "xx" FORCE)
+endif()
+
+
 if (BUILD_QCBOR_WARN)
     # Compile options applying to all targets in current directory and below
     add_compile_options(-Wall -Wextra -Wpedantic -Wshadow -Wconversion -Wcast-qual)