Clarify that those can't throw, preparing for the next commit.

Previously, a bogus value (whatever happened to be still in
`start_time`) was used.

See code comment.

Closes https://github.com/MusicPlayerDaemon/MPD/issues/629

This was missing in commit 0f488dce

Field 2 is called <released>, formerly used as <copyright>[1][2]. It is
formatted <year><space><company or author or group>, where <year> may be
<YYYY>, <YYY?>, <YY??> or <YYYY-YY>, for example "1987", "199?", "19??"
or "1985-87". The <company or author or group> may be for example Rob
Hubbard. A full field may be for example "1987 Rob Hubbard".

This change splits the <released> field at the first <space>, to retain
the <year> part.

The 51823 SID files in High Voltage SID Collection (HVSC) version 71
have the following distribution of dates:

    333 19??         11 1990-92       6 1995-99       2 2006-08
    827 198?         88 1990-93    2140 1996        530 2007
     32 1982         69 1990-94       9 1996-97      15 2007-08
      1 1982-83      49 1990-95       2 1996-98       2 2007-09
    255 1983       3467 1991          5 1996-99       1 2007-10
    677 1984         75 1991-92    1840 1997        430 2008
    775 1985         65 1991-93       4 1997-98      23 2008-09
      3 1985-86      10 1991-94    1276 1998          1 2008-12
     10 1985-87      35 1991-97       4 1998-99     631 2009
    943 1986       3320 1992        865 1999          1 2009-10
     12 1986-87      26 1992-93      24 200?        645 2010
      5 1986-89      59 1992-94     590 2000          1 2010-12
   2083 1987          1 1992-96       4 2000-01     538 2011
     31 1987-88    2996 1993        727 2001          1 2011-12
     44 1987-89      42 1993-94     875 2002        651 2012
   2510 1988         12 1993-95       2 2002-04     811 2013
    129 1988-89       2 1993-97     844 2003        790 2014
     91 1988-90    2737 1994          3 2003-05     740 2015
     58 1988-91      16 1994-95     842 2004        792 2016
   3466 1989         20 1994-96       2 2004-05     775 2017
     95 1989-90      17 1994-97     707 2005        638 2018
    150 1989-91    2271 1995          1 2005-06     284 2019
   1077 199?          2 1995-96       2 2005-07
   2834 1990          4 1995-97     785 2006
    119 1990-91       2 1995-98       6 2006-07

References:

[1] https://www.hvsc.c64.org/download/C64Music/DOCUMENTS/SID_file_format.txt
[2] https://hvsc.c64.org/info

High Voltage SID Collection (HVSC) metadata fields are encoded in
windows-1252, as described in DOCUMENTS/SID_file_format.txt:

https://www.hvsc.c64.org/download/C64Music/DOCUMENTS/SID_file_format.txt

If utf-8 transcoding fails, or the ICU library is unavailable, fall
back to plain ASCII and replace other characters with '?'.

The song length was previously undetermined.

Without this, the compiler may optimize accesses away.

Version 10.2+0.93+1 was released five years ago in 2014 and is the
first version to feature cdio_cddap_free_messages().  There is no way
to check the libcdio-paranoia version at compile time, so let's just
remove support for older versions instead of attempting to fix the
cdio_cddap_free_messages() check at build time.

Closes https://github.com/MusicPlayerDaemon/MPD/issues/613

This reverts commit f7ed7446.  It was
a bad idea, because MAD_F_MIN and MAD_F_MAX do not represent the
clamping limits, but the theoretical minimum and maximum values of the
mad_fixed_t data type.

Closes https://github.com/MusicPlayerDaemon/MPD/issues/617

The check IsSeekableCurrentSong() was added by commit
44b20024 in version 0.20.19, but it
caused a regression: by doing the branch only if the current song is
seekable, the player would restart the current song if it was not
seekable, and later the initial seek would fail; but we already know
it's not seekable, and so we should fail early.

libmad has a hard-coded maximum PCM buffer size; if we make our
output_buffer just as large, we can avoid the loop, because any
possible size will fit.

These will not be used until they are initialized in SyncAndSend().

Apparently, libmad not only inserts 529 samples of silence at the
beginning of the file, but also removes them at the end.

This solves the last piece of
https://github.com/MusicPlayerDaemon/MPD/issues/601

Closes https://github.com/MusicPlayerDaemon/MPD/issues/601

libmad requires padding the input buffer with "MAD_BUFFER_GUARD" zero
bytes at the end of the file, or else it is unable to decode the last
frame.

This fixes yet another bug which prevented this plugin from decoding
the last frame, see
https://github.com/MusicPlayerDaemon/MPD/issues/601

The Xing/LAME frame indicates how many frames there are, but that
excludes the initial Xing/LAME frame.  Therefore, it should not be
counted.

This fixes an off-by-one bug which caused the last frame to be
skipped, fixing one part of
https://github.com/MusicPlayerDaemon/MPD/issues/601

Increment "current_frame" after processing the frame.

This also fixes a bug which caused the bit rate to not update after
seeking.