Name the scan lock file from the database's resolved path (closes #101)

The lock file was the configured database path with .lock appended, so
a scan reaching the database through a symlink to it locked a different
file and ran alongside a scan holding the lock through the real path.
The lock file is now named from the database's path with symlinks
resolved. Before the database exists, its directory is resolved and the
file name joined to it, so the error names the same lock file either
way. testDBPath resolves its temp dir, so lock errors name the expected
path where the temp dir is itself reached through a symlink.

Model: opus-5-5

TODO.md

@@ -29,6 +29,10 @@
 
 # Completed Steps
 
+- the scan lock file is named from the database's path with symlinks resolved,
+  so a scan through a symlink to the database fails while another scan holds the
+  lock (2026-10-14, https://git.eeqj.de/sneak/sfdupes/issues/101)
+
 - `scan` records mtime to the nanosecond, as whole seconds in `mtime` plus
   `mtime_nsec`, and compares it at that resolution, so a same-size rewrite
   within the same second is re-hashed (2026-10-07,