frontends (tools/webrtc): add a options to merge video for PiP and to specify size: - the `merge_pip` option is now used to indicate if local feedback and remote video streams must be merged with local feedback being a PiP of remote video. By default, it's done for `SINKS_AUTO`, but it can be manually specified to use it with `SINKS_APP` or to split ``SINKS_AUTO`` in 2 windows. - target size of the compositor used with `merge_pip` can be specified. By default, 720p is used. rel 428

syntax: glob
bridge_constructor/generated
.build
build
ctags_links/
dist/
.eggs/
env/
libervia_backend.egg-info/
MANIFEST
.mypy_cache/
*.orig
*.pyc
*.pyo
.pytest_cache
*.pyv
Session.vim
stubs/
*.swo
*.swp
tags
tests/e2e/report*
_trial_temp/
twistd.log
twistd.pid
twisted/plugins/dropin.cache
*.un~
.vim
.vscode