opendir need to check that the given path is actually a directory
before to open it.

Moreover now open can be called with the O_DIRECTORY flag in order to
open an existent directory

Support for the following calls has been added:
 - opendir
 - readdir
 - closedir

The readdir call use the new RPC get_dirents that ask to all the
nodes (broadcast) about all the existing first-level entries
of a specific directory.

Allow to add already instantiated OpenFile to the OpenFileMap.

This will come in hands when we will have sub-classes of the OpenFile.
In that case this new add function will accept all object that are
derived from the OpenFile class. See directories functionalities.

The new PreloadContext class is used as singleton to store the general context
information of the interception library.

The plan is to move global accessed variables into this sigleton.

At the moment just the logger object has been moved into it.

It is only relevant for sparse files but the current implementation
has a too high overhead on a single node (~10%). We need to find
a better solution at a later point.

This is an optimization for mdtest which shows very slow performance
in removing files, getting worse with an increasing number of nodes.

Postpone the creation of the OpenFile object until we are sure that the
file actually exists and we have enough access rigths for it.

Since POSIX.1-2001 the S_IFMT (0170000) bitmask of the file mode must
contain the file type.

At the moment we correctly support
 - [S_IFREG] regular file
 - [S_IFDIR] directory

Instead use margo_iforward. Also only two bulk_handles are now used:
1 for margo_ipc and one for margo_rpc RPC requests.

The LRU map caused multiple issues with memory mappings and caused
severe errors and hanging situations. It has been replaced with a
std::map.

An Argobots mutex was added for rpc address lookups as many threads
might insert and lookup this map in parallel. This caused bulk_transfer
seqfauls as well as other file system crashes.

Cleanup of file system logging.

Only use 1 segment now but with multiple transfers

We need to be explicit about I/O as it cannot be simply repeated. A response
from the daemon will come eventually, either success or failure.

Previously, an ES was created for each destination in each read or write.
Creating and freeing these ES is costly. Also, these ESs had separate pools,
which is bad practice. Multiple ESs work on a single pool. The ABT_sched
decides which ULT or Tasklet is run by which ES at any time.

This change does also reduce CPU load as library IO RPC calls used CPU time
while waiting for the response.

ADA-FS's statfs will fill the struct mostly with rough estimations.
It is not supposed to return accurate data but "some" data for applications to work

Added lseek64 syscall and used file flags properly