* changes:
  LeAudio: Enable essential audio context logs
  LeAudio: Remove Lc3 spec. code from client_parser
  LeAudio: Offloader related code encapsulation Part 2
  LeAudio: Offloader related code encapsulation Part 1
  LeAudio: Correct header inclusion chain for le_audio_types.h
  LeAudio: Improve BidirectionalPair type
  LeAudio: Improve data path setup
  LeAudio: Bind configuration provider life time with the Codec Manager
  LeAudio: Fix not releasing the codec resources
  LeAudio: Introduce codec wrapper interface

* changes:
  Floss: fix typo of Lc3Param
  Floss: define MMC DBus service proto
  Floss: define MMC interface

This will improve triaging any potential context type related
issues when debug logs are not enabled.

Bug: 295972694
Test: atest bluetooth_le_audio_client_test
Test: manual
Tag: #feature

Change-Id: Ia591e1d32c2efcdef0c29312f316d6004f03ae09

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test  bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: Iec29b128a97fe013d44be94c6c34e282cfe83279

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test  bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: I772e238c11d5b821b0b49ab1bad0891c551d6a2e

This will be needed for the multi-codec support.

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test  bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: I785a0f697ca49ec8a54cb7c7ebf3574e03530b90

This file is included by multiple modules and should not include
any redundant headers, adding more dependencies than needed.
The other modules should rather directly include headers for the
API they use.

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test  bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: Iaa1876ff6279fa2d979c65fdb2182a405d395a2b

 - Changed get() to follow the common idiom, while copy constructor
will be called anyway if non-reference type is on the left hand side
of the assignment. There is no need for explicit reference getter.

 - It no longer requires comparison operators, which may no be defined
for the more complex types.

 - For the more complex types a single bidirectional value cannot be
 computed (e.g. a bidirectional pair of ASEs).

This allows us to wrap more complex types and simplify the code further.

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test  bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: I35518479075732d8aa87082dd9267864161f43b3

The data path configuration should be provided by the audio set
configuration provider. For now, these are based on the current
codec location (either HOST or ADSP). In future this should
be provided from the vendor interface over the AIDL.

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: If39a1d1ac089f236e4ed9258644c197a57f1e723

In the near future, the Codec Manager, will become the main source for
the audio set configurations.

Bug: 295972694
Test: atest bluetooth_le_audio_test bluetooth_le_audio_client_test bluetooth_test_broadcaster bluetooth_test_broadcaster_state_machine
Change-Id: Id779e17385aef8e675794a90726329e72c0de6f4

After stream gets suspended and the active group goes to IDLE state, we
should be cleaning up the codec. It gets created when the stream is
resumed.

Bug: 295972694
Bug: 262481440
Test: manually - start, stop stream and wait, then resume and observe no warning about the encoder instance not being destroyed
Change-Id: I781e603e0453984841821197c139f2b8927173bf

This extracts the encoding logic from client.cc,
for better code separation and maintainance.

Bug: 295972694
Bug: 262481440
Bug: 262481609
Test: atest bluetooth_le_audio_client_test bluetooth_le_audio_test bluetooth_test_broadcaster_state_machine bluetooth_test_broadcaster
Change-Id: Iee2d1a53c0a0955a8ff7bdd0e9e705ad17159cd9

Bug: 296819230
Tag: #floss
Test: Manually register the same UUID twice, the later is rejected
Test: mmm packages/modules/Bluetooth
Change-Id: I0421c808c8d165f9b56775cfbc32b02b04911582

Bug: 296912033
Test: atest AvrcpControllerStateMachineTest
Change-Id: Iab36d774f540d3f5eded33c98432dd6a3e41f402

Merge "Remove 'asynchronously_start_l2cap_coc' flag since it has been enabled in prod bug-free since tm-qpr2-release." into main

Bug: 290844784
Test: m .

Change-Id: Idb1017560d229064056be8d9aab7a4018c0b4719

Merge "Remove 'queue_l2cap_coc_while_encrypting' flag since it has been enabled in prod bug-free since tm-qpr2-release." into main

* changes:
  Stop using BluetoothA2dp and BluetoothHeadset Broadcasts for CONNECTION_STATE_CHANGED in SilenceDeviceManager.
  Stop using HFP active device changed broadcast for SilenceDeviceManager.
  Stop using A2DP active device changed for SilenceDeviceManager.

Merge "Remove dependency on BT broadcasts inside the Bluetooth stack for Volume control connection changes." into main

Merge "Posting the connection change on a handler to make sure that we don't enter a stack overflow. (If connection disconnects we also remove state machine which is a conflict)." into main

Bug: 296085333
Change-Id: I5eecb4f8f000877e22227f99a3780f896573c739
Test: Owners files

Bug: 296259098
Tag: Floss
Test: mma -j80
Change-Id: I8f590efb6cec762584c2e782ad32df15bc882295

If CIS got disconnected due to timeout and ASE goes to QoS configured as
per specification, Initiator should be able to re-enable the stream on
this ASE.

For this reason, we keep ASEs as active to be able to restart the
stream.

Bug: 296818373
Test: atest bluetooth_le_audio_test
Tag: #feature
Change-Id: Ifc7cf29ff11d6dbf0566d03985f3af07a20588df

When disconnecting a peer device, we first disconnect all of the
connected profiles. However, if the profile connection is created via
a listening/connecting socket, floss layer doesn't know the current
state of the connection, since the corresponding file descriptor has
been passed to whatever client that asks for the socket.

Therefore, add a disconnect_all method to the socket interface to ask
libbluetooth to disconnect all socket connections which corresponds
to a particular device address.

Bug: 288939390
Tag: #Floss
Test: Fast pair Pixel buds pro, disconnect from peer, then press
      disconnect from UI about 2-3 secs later. Verify the FP profile
      is disconnected so the baseband connection is also disconnected.
Change-Id: I5927fdd4d2a99240af5f007564b0f5b42136c869

Remove 'queue_l2cap_coc_while_encrypting' flag since it has been enabled in prod bug-free since tm-qpr2-release.

Bug: 290845993
Tag: #stability
Test: m -j $(nproc) ; Android should build without the flag as well.
Change-Id: Icb19dfe9ca8d36bb9e95aeb477455adabc486e1c

Remove 'asynchronously_start_l2cap_coc' flag since it has been enabled in prod bug-free since tm-qpr2-release.

Bug: 290842781
Tag: #stability
Test: m -j $(nproc) ; Android should build without the flag as well.
Change-Id: I43293d3c7c45c2aab0e7c29304b86045b727ffb4