This is instead of caching it inside nilmdb.server.wsgi_application.
Might make things work a bit better in case the web server decides
to call wsgi_application multiple times.

This ensures that the same Python sub-instance handles the request,
even if it's coming in from two different virtual hosts.

This way, errors like "database already locked" can be fixed and the
page reloaded, without needing to restart Apache.

Otherwise we might inadvertently catch SystemExit or KeyboardExit or
something we don't want to catch.

This isn't necessary for correct behavior: if the database is killed,
the old flock() will go away when the file descriptor gets closed.

When enabled, lines like "# interval-start 1234567890123456" and "#
interval-end 1234567890123456" will be added to the data output.  Note
that there may be an "interval-end" timestamp followed by an identical
"interval-start" timestamp, if the response at the nilmdb level was
split up into multiple chunks.

In general, assume contiguous data if previous_interval_end ==
new_interval_start.

Added new flag "-R" to command line to perform an automatic removal.
This should be the last of the ways in which a single command could
block the nilmdb thread for a long time.

Server class will retry as needed, as with stream_extract and
stream_intervals.

Instead, always take the subset, which creates a new interval.
Also adds a small optimization by moving the 'if orig' check outside the
loop.

Clients might need to to Interval math too, so move a simple Interval
class and start putting helpers in there.

This lets us use simple indexing to get timestamps from the table,
which allows us to use 'bisect' directly without needing a proxy class.