If an early exception gets caught (e.g. from
AllocatedPath::FromUTF8Throw()) before
DecoderControl::CommandFinishedLocked() is called, the decoder thread
would go in an endless loop, because DecoderCommand::START is still
set.

Closes #118

This eliminates some overhead, because the compiler doesn't need to
consider these functions throwing.

This repairs the damage to commit 74dbaade done by commit
b3f5b493

 https://bugs.musicpd.org/view.php?id=4595

Catch ENOTDIR exceptions and in that case, try all matching decoder
plugins which implement the "container_scan" method.

 https://bugs.musicpd.org/view.php?id=4561

Replaces decoder_data() and others.

Code simplification.

Be exception-safe.

Prepare full C++ exception support in the decoder thread.

Fixes memory leak after stream failure.  See
https://bugs.musicpd.org/view.php?id=4562

This reverts commit 995cd954.  It was
a bad idea because it caused assertion failures when a decoder plugin
aborted with an exception.

Allow decoders to throw std::runtime_error.

Fixes an assertion failure in tag_ape_scan() which is because
decoder_input_stream_open() returns an "unready" InputStream.

Adds support for stream codecs which havn't been explicitly listed in
ffmpeg_mime_types.

Keep only the last error.  This fixes bogus aborts due to the error
check in decoder_get_virtual_command().

decoder_load_replay_gain() will create the InputStream anyway, so
we're not saving any overhead by opening the InputStream on demand
only.