LogFileCollector exception is misleading when Oracle ARC is slow to archive

When the Oracle ARC process is slow to archive a log, this often leads to the message:

> None of the log files contain offset SCN: , re-snapshot is required.

We recently fixed the _deleted log_ use case in #1532; however, we did not update the last exception message to provide more context on why the log could not be found, since the goal of #1532 was to differentiate deleted logs from missing ones.

I propose the log message to shift to:

> None of the log files contain offset SCN: after {} attempts. This could be due to a slow archiver process in the database. Please consider raising {} (the default is {}) if this continues. If a connector restart does not eventually solve the issue, and the expected log can no longer be resolved, a re-snapshot is required.

This informs the user that we retried X times, directs them to the internal property to increase it if the issue persists, and notes that a re-snapshot might be required if the issue persists because a log file is in an unexpected (non-deleted) state.