Bscampp

1. Implemented functionality to recompile C++ codes locally so that dynamic
libraries are referred to correctly.

1. Changed from using original sequence names from query input to using
pseudo names (indexed from 1 to N, N the number of queries). This is to prevent
issues with nontypical characters in sequence names, such as ":".
2. Added better parallelization for the placement phase. Now jobs can run
in parallel to fully take advantage of more cores. By default, each job can
use 2 cores (controlled by `--cpus-per-job`). The number of workers
that can run in parallel is determined by `max(1, num_cpus // cpus_per_job)`.

1. Supported .gz format input files for alignment.
2. Included all other changes made in v1.0.4a.

1. Removed redundant parameter `--model` since the user-provided info file
is sufficient to run `pplacer` or `epa-ng`. Also this means that it is
not needed to distinguish between nucleotide/protein sequences as long as
the user provides the files correctly.
2. Added a new test for using BSCAMPP on protein sequences, see
`examples/run_bscampp_prot.sh`.

1. Version bump to avoid the entire "pre-release" tag with beta versions.

1. Added SCAMPP funtionality and its binary executables.