FS-10989: [core] Fix CUSTOM_HASH hash table race in switch_event logic causing crashes during bind to events.

Merge pull request #10 in FSA/freeswitch-advantage from ~ANDYWOLK/freeswitch-advantage:bugfix/FS-11020-add-missing-modules-to-the-msi-installer to v1.8-fsa

* commit '22a2c0af':
  FS-11020: [Build-System] On Windows: Add missing modules to the msi installer, fix mod_gsmopen build, remove mod_skyopen, disable libav and libx264, cleanup.

FS-11020: [Build-System] On Windows: Add missing modules to the msi installer, fix mod_gsmopen build, remove mod_skyopen, disable libav and libx264, cleanup.

Merge pull request #9 in FSA/freeswitch-advantage from feature/FS-11063-dtls-state-setup to v1.8-fsa

* commit '4a59554d':
  FS-11063 Use compile time constants in dtls_state_setup

gcc 7 complains about possible format truncation:

mod_cdr_mongodb.c: In function 'my_on_reporting':
mod_cdr_mongodb.c:242:45: error: '%d' directive output may be truncated writing between 1 and 10 bytes into a region of size 4 [-Werror=format-truncation=]
   snprintf(idx_buffer, sizeof(idx_buffer), "%d", callflow_idx);

The char * idx_buffer has a size of 4 Bytes, and according to gcc's
calculation it is possible that up to 11 Bytes might be copied into it
via int bson_idx.

This commit adds an extra 8 Bytes to char * idx_buffer, which silences
the warnings.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>

Add the RTT field to the RECV_RTCP_MESSAGE event emission which allows external listeners to compute MOS RTCP from the event.

This reverts commit c093b2d3.

RTCP/Receiver Report/lost field is a _signed_ 24bit integer and it could be negative (in case of UDP duplication)
+ any negatives now threated as huge uint32_t
+ set this field properly on __BIG_ENDIAN
+ correctly read this value in received RTCP on all arches

Conflicts:
	libs/sofia-sip/.update

FS-10908: [mod_amqp] AMQP routing key (format_fields) formation broke and invalid reads reported by valgrind #resolve

Ethernet packets.

Memory leak happening during reconnection attempt after connectivity loss with
RabbitMQ broker.

With these changes, memory leak issue has been addressed and the object has
been properly freed.