Cardano Updates

Signed-off-by: Chris Gianelloni <[email protected]>

Signed-off-by: Chris Gianelloni <[email protected]>

Signed-off-by: Chris Gianelloni <[email protected]>

This is a big and unpleasant commit; but vastly necessary. Multiple
things are going on, but before I explain further: why are we doing
this?

We have decided to move away from Pallas. There are various reasons
motivating this and a proper EDR will come ASAP. The gist being that we
ended up working more _around_ Pallas than _with_ Pallas. The amount of
duplication and workaround have reached a point that makes this switch a
lot more painful than I'd have liked.

In short, here's what's happening:

- The kernel top-level `lib` has been broken down into many smaller
  modules; the general strategy being "one type per module", with
  associated methods and traits.

- This shall allow us to now migrate in a more incremental fashion from
  Pallas to Amaru's own kernel / prelude. And, importantly, to give more
  thoughts about the public API we want to expose; for us, but also for
  downstream consumers.

- Few methods have been moved around, from the kernel to other places or
  vice-versa when it was relevant.

- Many duplications were already addressed; sometimes requiring a bit of
  finesse, renaming and moving types around. Some heavy work remain in a
  couple of places such as the amaru-plutus & the script context, as
  well as the "memoized" sections, which needs to be promoted as the new
  defaults.

Signed-off-by: KtorZ <[email protected]>

Signed-off-by: Chris Gianelloni <[email protected]>

Signed-off-by: wcatz <[email protected]>

Signed-off-by: Chris Gianelloni <[email protected]>

topology

Track checksums of block producer keys and topology values as pod
  template annotations. This allows Kubernetes to detect changes to
  secrets and configmaps that would otherwise be invisible since the
  resource names don't change between upgrades.

  Users can control rollout behavior per release via updateStrategy.

  Closes #327

This is a big and unpleasant commit; but vastly necessary. Multiple
things are going on, but before I explain further: why are we doing
this?

We have decided to move away from Pallas. There are various reasons
motivating this and a proper EDR will come ASAP. The gist being that we
ended up working more _around_ Pallas than _with_ Pallas. The amount of
duplication and workaround have reached a point that makes this switch a
lot more painful than I'd have liked.

In short, here's what's happening:

- The kernel top-level `lib` has been broken down into many smaller
  modules; the general strategy being "one type per module", with
  associated methods and traits.

- This shall allow us to now migrate in a more incremental fashion from
  Pallas to Amaru's own kernel / prelude. And, importantly, to give more
  thoughts about the public API we want to expose; for us, but also for
  downstream consumers.

- Few methods have been moved around, from the kernel to other places or
  vice-versa when it was relevant.

- Many duplications were already addressed; sometimes requiring a bit of
  finesse, renaming and moving types around. Some heavy work remain in a
  couple of places such as the amaru-plutus & the script context, as
  well as the "memoized" sections, which needs to be promoted as the new
  defaults.

Signed-off-by: KtorZ <[email protected]>

  topology

Track checksums of block producer keys and topology values as pod
  template annotations. This allows Kubernetes to detect changes to
  secrets and configmaps that would otherwise be invisible since the
  resource names don't change between upgrades.

  Users can control rollout behavior per release via updateStrategy.

  Closes #327

* add Algolia site verification tag

* Add Algolia search

* add trailingSlash: true

* Hide search bar

* fix links

* Add modal search

* Update glossary layout to work better with search

Fix nightly CI