The `rp_manager_set_storage_credentials` is now skipped when running a
trace in `revng trace run`. This is needed because the storage backend
when running a trace is replaced with a local one and calling the
function does not make sense.
On the C++ side of revng it is important that LLVM's signal handlers are
registered within the initialization of `revng::InitRevng`, as this
allows downstream users (e.g. `revng daemon`) to preserve its signal
handlers before setting up the C++ infrastructure. In
`Tracing/Runner.cpp` there's `CommandHandler` which is statically
initialized and called `llvm::sys::AddSignalHandler`, this happened at
`dlopen` time, breaking the promise outlined above. Rework the code so
that the function is called lazily instead of statically.
Add support for tracing onto the PipelineC. This is done by:
1. Creating wrapper functions for each PipelineC function with the
script in `scripts/PipelineC_add_tracing.py`. These will call a
special function called `wrap` which will ultimately call a method
with a `_` prepended to the name
2. Conversion of all PipelineC methods in `PipelineC.cpp` to `static`
and their rename with a `_` in front, in order for them to work with
the wrapper function in (1)
3. Generation of 2 additional include files, one for types and one for
functions, to be used by users of tracing files in order to have
introspection.
These steps allow the creation of a trace file with the use of the
`REVNG_C_API_TRACE_PATH` environment variable. The traces can then be
used in conjunction with the `revng trace run` and `revng trace
inspect` commands.