Provide better status code conversion from OS_GetResourceName

Correct code that was not correctly using the CFE_SB_ValueToMsgId or
CFE_SB_MsgIdToValue conversion wrappers where required to do so. This
should be used whenever the value is intentionally converted to/from
an integer.  The CFE_SB_MsgId_t type should not be assumed to be
an integer in nature.

Historically the CFE_SB_MSG_ID_INVALID constant was defined as 0xFFFFFFFF/-1,
where 0 was considered valid.

Although 0 is indeed valid for the first word of a CCSDS primary spacepacket header,
it is not actually valid for use with SB.  Because SB is now more decoupled from CCSDS
header definitions, there are a number of advantages to using 0 instead of -1, as
it is more passively safe:

- Objects which are cleared as part of normal BSS clearing will be set invalid
- Objects which are memset to zero will be set invalid

In contrast, when the invalid value is nonzero, objects which are memset/cleared
are valid by default, and must be actively set invalid to be safe.

This makes CFE_SB_MsgId_t to be a safe wrapper around CFE_SB_MsgId_Atom_t,
such that the values cannot be silently/implicitly interchanged with other
integers.

This enforces that the MsgId/Value conversion helpers must be used when
conversion to/from integers is intended.

Use a conversion macro, rather than directly referencing a member name/heirarchy
within the CFE_MSG header structures, when calling SB APIs that accept
a CFE_MSG_Message_t* type.

The conversion macro is specific to how the headers are actually defined.

Adds a new module called "config" that tracks simple
key/value pairs.  All values are const.

As a proof of concept, all version information is mirrored
into this key store, and ES is updated to use this instead
of the GLOBAL_CONFIGDATA when generating events.

Use an actual `void*` value when invoking functions
that output an opaque pointer value.  If necessary
the value of this pointer can then be assigned to
the real pointer type to dereference.

Utilize the common setup/wrapper functions in ut_support.c whenever
possible MSG dispatching in the generic handler.

…ME-ClockState-Enum-t-in-cmd/tlm-and-handling

Fix nasa#1546, use cfe time clock state enum t in cmd/tlm and handling nasa#1663

Fix nasa#245, nasa#1944, Message ID type improvements

Fix nasa#1984, avoid aliasing warnings

Fix nasa#1300, CFE_ES_MemAddOff structure to associated CFE_ES_MemAddress_t
and CFE_ES_MemOffset_t

Adds CFE_SB_RequestToSendEvent/CFE_SB_FinishSendEvent wrappers around
all events generated by CFE_SB_TransmitMsgValidate.  This is avoids
the potential for a recursive loop if configured improperly.

Fix nasa#1952, patch for recursive event loop

Fix nasa#1672, return type conversions in CFE_ES_GetTaskName

…types

Fix nasa#1818, EVS functional test add for different event types

Fix nasa#1988, consolidate repeated MSG stub setup in sb_UT

…unction-protos

Fix nasa#1319, Remove extern from function prototypes

Re-adds externs removed from CFE_ES_ObjectTable and CFE_ES_Global to
fix RTEMS build

…gistry

Fix nasa#1979, implement abstract config registry module

Fix nasa#1965, better message type pointer conversion

- Uses CFE_PLATFORM_SB_DEFAULT_MSG_LIMIT

…d_msg_limits_in_TIME_services

Fix nasa#1537, replacing hardcoded message limit in TIME services

…_IsNewOffset_only

Fix nasa#1928, checking against IsNewOffset only nasa#1999

…zero-compare

Fix nasa#1302, Use macro in CFE_ResourceId_IsDefined

- Set new build baseline for cFS-Caelum-rc4: v7.0.0-rc4
- Standardize development indicators in cfe_version.h to use 0xFF in
MissionRev to indicate development version