restore the ll[io]fstream because we need them as wrappers on Windows for wide char paths; on other platforms they are now just typedefs to the std classes

replace llinfos, lldebugs, etc with new LL_INFOS(), LL_DEBUGS(), etc.

The logic in llcommandlineparser.cpp's setControlValueCB() callback function
for converting command-line switch argument values from string to the actual
type of the map-to settings variable had a couple special cases for boolean
and LLSD array. But for S32, U32 and F32, it simply used default LLSD
string-to-numeric conversion. LLSD's string-to-numeric conversion is a bit
lame: for non-numeric strings, it shrugs and returns 0.
Introduce onevalue() and badvalue() helper functions that, like certain errors
during command-line parsing, throw LLCLPError. Use them to streamline certain
redundancies in setControlValueCB(). Introduce convertTo<T>() helper function
that uses boost::lexical_cast() for slightly more stringent conversions. Add
cases for U32, S32 and F32 targets.
setControlValueCB() is actually called only by LLControlGroupCLP::notify(),
not during actual command-line parsing.
Make LLControlGroupCLP::notify() return bool. Make it catch LLCLPError, set
the error for getErrorMessage() and return false on that exception.
Package LLAppViewer::initConfiguration()'s response to initParseCommandLine()
returning false as a new handleCommandLineError() function; invoke it both
there and when LLControlGroupCLP::notify() returns false.

A small, fixed set of cmd_line.xml switches can't reasonably be mapped to
settings variables, mostly because they affect the settings machinery itself.
Other than those, every new cmd_line.xml switch should map-to a settings
variable. Validate that only the known set does not have map-to; validate that
map-to variable actually exists.

Lacking time to properly test new LLStringUtil::getTokens() against the
present (different!) command-line scanners in LLExternalEditor::tokenize() and
LLCommandLineParser::parseCommandLineString(), just annotate as future work
the goal of unifying them... SIGH.

Re-applied fix originally committed in 1e93695ded30
Reveiewed code, appears straightforward.

VWR-23459 FIXED Viewer compiled against Boost-1.42 crashes when certain command line options are given

used Aleric's SG2 changeset from http://svn.secondlife.com/trac/linden/changeset/3600
	patching file doc/contributions.txt
	Hunk #1 succeeded at 73 with fuzz 2.
	patching file indra/newview/llcommandlineparser.cpp
	Hunk #1 succeeded at 268 with fuzz 1 (offset -8 lines).

Edited doc/contributions.txt to create an entry for Aleric and moved the
issue ID there (patch wasn't able to place it at the right position,
lacking any context).

Result of svn merge -r107256:107258 svn+ssh://svn/svn/user/phoenix/license_2009_merge into trunk. QAR-1165

dataserver-is-deprecated

merge release@88802 Branch_1-20-Viewer-2-merge-1@89178 -> release

merge Branch_1-20-Viewer -r 84432 : 84760 -> release

QAR-369

QAR-369 - viewer-cleanup2-7 81916 merged into release.