LLIIA ETL Updates

Synchronize upstream with ETL Updates

The finalist attachments are now added without revealing information
about their names / proposal IDs.  This is following the same pattern as
Climate2020.

We don't need the finalist tabs after all in this competition.

These tabs are deployed but have not yet been added to the repository.

This updates the modern competitions to use the common deploy tooling,
which includes a decrypt helper function and the help documentation that
explains how to call the script.

There are a few legacy competitions which have not been updated, but
potentially could be in future.

These two scripts will be leveraged by the various competition deploy
scripts.  Currently this logic is duplicated across several
competitions.

Just adding the GlobalViewMediawiki column and updating the shas to
checkout.

Also included a template change where Attachments was forgotten

These are:
* Adding a second sheet information adder, for when a second spreadsheet
  needs to have some information added to the overall dataset
* Allow MultiLineProcessor to specify the string it's split on
* Allow tocs to take in multiple columns to build up the groupings on
* Output a message in the case that adding a page fails for whatever
  reason

Just copied over from the template and replaced all the variables.

This should be a complete shell for a new competition, with some of the
documentation moved out of the base README.md.  With these instructions,
someone should be able to quickly get a new site up and running.

This includes:
* Reverting changes to LoneStar as it's not the first competition going
* Changing LLIIA2020 from GV, since this is going live
* Parameterizing the torquedataconnect role's git checkout so old
  competitions are broken by this
* Other small whitespace fixes

Change things from GV to regular because we're no longer demoing.

Global View needs to come along for the ride with the ansible changes.

Also required reversing the wiki config list as needed by the update.
See README in torque data connect for more information.

For proposals that come from other competitions, there's normally a
<teamcomments /> tag, that needs to render in Global View to the empty
string.

This required a few small changes to LoneStar and LLIIA, in that the
competitions needed to be renamed to GVLoneStar and GVLLIIA, so that
they could exist on the production server alongside their non Global
View enabled counterparts.

Otherwise, this large change builds out the initial deployment of Global
View, including:
* The GV ansible config
* The GV pipeline
* The GV enablign etl changes to LLIIA and LoneStar
* The base etl changes to add common Global View data
* Update to torquedata for the new deployment of the django enabled
  torquedata

I hope I didn't just cause problems, but we were being
more-specific-than-necessary when copying pdfs over to the attachment
directory.  This now copies everything, which is handy because we
changed one of the appendixes to `Overview` and this allows us to copy
it without adding another line.

Accidentally removed the `:` from an existing deploy parameter, which
made everything very sad and upset.

The overview appendix item is being moved / renamed to be a standalone
document.

The new first tab is "Overview" which will host the 2 page overview.

The prospectus still exists, it just isn't the first page any longer.

We decided to change it once again to "Analysis"

It's now named "Diligence".  The move will also need to be manually
performed on production.

Issue #97

The new tab is now "Synthesis".  This means a change to the name, the
title, and the slug.

In order to do this we will need to manually perform the move on
production.

Issue 97

The content of appendices is being added to the main / prospectus tab.
This move will need to be manual for now (and has already occurred in
production)

Issue #97

This modification allows templates to be populated using envsubst. This
change is similar to the one made for the ansible templates, and the
instructions to use it are identical to the example already contained in
README.md

Issue #95

There was a potential bug where simplesaml autoconfig would populate
with an empty '' => '' if the okta variables were not set.

This adds a conditional, skipping that step if there is no value
present.  There are still potentials for error in the event that
simplesaml is never provided any value, however, as at least one
authentication endpoint must be defined for simplesaml to work.

Issue #93

The 100Change2020 deploy script supports the downloading and decrypting
of key competition data before running the ingestion. This could not be
scripted in non-interactive environments because there was no support to
pass in a GPG passphrase.

This adds a new `-g` flag to deploy, which accepts a passphrase.

We will likely want to support this flag in other competition deploy
scripts, and it may be valuable to consider creating a set of common
utility methods imported across the various deploy scripts to avoid DRY
violation.

Issue #93

Mediawiki sometimes needs the php-mysql task to run.  It doesn't really
hurt to proactively include it here (we already do so for psql).

The alternative would be to remove database dependency installations
from the mediawiki task and require that the calling role determine
which dependency to include, but that is a larger scope concept and not
necessarily something we would prefer anyway.

Issue #93

There were two typos in the 100Change2020 document.  One was a full
blown typo, but the other was just a divergence from the standard of
mapping the environment variable to the variable name that is being
populated, (whenever the variable name is specific enough to not be
ambiguous)

Issue #93

Fixed one small mistake and added a handy reference point for people who
might not have read through the DESIGN.md but want to confirm their
installation is working via login.

Ansible will, by default, overwrite permissions for a mysql user when
setting new permissions.  This modifies the task so that it enables
`append` mode. That way if you are re-using names you won't find
yourself unhappily and unexpectedly lacking permissions.

Issue #85

Also includes a fix because some of the spreadsheets were
malformed, with a lot of extra columns.

They had a trailing slash, which actually breaks the etl pipelines.
This was confusing for any new developers, and so should be changed to
be accurate.