GitLab

  - Clarify what we learned in experimentation
  - Clarify N-EWMA and next_cc_event behavior
  - Clarify order of operations in BDP estimator
  - Clarify and expand on parameter tuning
  - Clarify Vegas and Nola descriptions
  - Mention that we need to test intermittent clients
  - Update old trac URLs

Since 332-ntor-v3-with-extra-data.md was faster than us

- Mention that an extra hop is added on service-side intro circs
- Add circuit path length diagram
- Mention max number of L2 guards
- Add control port events to control spec

This is never necessary, since we only count circuits who have
completed at least one hop.

Closes #57

 - Correct algorithms to update once per cwnd
 - Clarify orconn blocking and edge connection checks
 - Mention Westwood RTT backoff, and a couple other fixes
 - Specify BDP estimators based on testing and bugfixing.
 - Specify BDP-based congestion algorithm (TOR_NOLA)
 - Update consensus parameters with tuning notes.

Closes #51

Signed-off-by: David Goulet <dgoulet@torproject.org>

Closes #50.

Signed-off-by: David Goulet <dgoulet@torproject.org>

We indicated that "0" meant "no more messages" but we didn't say
what happened if there was no space to put such a 0.

Closes #41

Closes #40.

Signed-off-by: David Goulet <dgoulet@torproject.org>

- Rename 'overload-reached' to 'overload-general'
- Simplify 'overload-ratelimits' for engineering reasons
- Add versioning on the extra-info fields
- Add a few more metrics

Originally designed in tor#25573 as part of a defense for the
DropMark attack by Rochet and Pereira.

Closes #33.

Signed-off-by: David Goulet <dgoulet@torproject.org>

Also clarify and improve wording of the timeout calculation section.