Open sourcing tasks

[woodruffw]

TODOs for a public release:

• Purge the repo of internal documents, build products, and old (SL1, 1.5) history.
  • Possibly just create a private repo with the original tree and force-push over this one?
• Copy the wiki over to the archive and wipe the one in this repo.
• Rewrite the README (README: Add diagram, begin cleanup #387).
  • Logo?
  • Visualization/graphic of SL2's overall structure?
• Set up a CI (CI builds and tests #386).
• Set up builds (CI builds and tests #386).
• Provide an initial downloadable and installable version.
• Clean up some of the harness code (we can probably automate some of this). (harness, gui: Autoformatting #382)
• Clean up the DR clients and SL2 server (we can just throw clang-format at these). (sl2: clang-format #383, sl2: clang-format #385)
• Freeze the database structure, or come up with a real migration process for updates.
• Mark easy outstanding issues as "help wanted"
• Standardize the DR client names (fuzz_dynamorio should be fuzzer, tracer_dynamorio should be tracer) (Rename fuzzer and tracer #388)
• Scrub references to fuzzkit and other pre-alpha names (Remove fuzzkit references #389).
• Build the DR clients/triage tools as Release, not Debug.
• Automatic cronbuilds, a la DynamoRIO
  • Travis's support for WIndows deployments is still a little immature, so this will probably have to wait for a bit. (Broken out as Weekly cronbuilds #390)

[ehennenfent]

If we go the migration process route, alembic is the preferred method for upgrading SQLAlchemy databases. IIRC you still have to be a little bit careful with your database design because it's possible to make schemas that are incompatible with one another.