Return modeling commands from KCL execution (#4912)

* Add Rust side artifacts for startSketchOn face or plane

* Add Rust-generated artifacts to ExecOutcome

* Add output of artifact commands

* Add new output files

* Wire the artifact commands to the artifact graph creation

* Fix to use real PartialEq implemented in modeling commands

* Fix modeling commands with zero fields to work

* Fix missing artifactCommands field in errors

* Change artifact graph to be built from artifact commands

* Wire up ExecState artifacts, but not using them yet

Co-authored-by: Kurt Hutten <k.hutten@protonmail.ch>

* Remove unneeded local var

* Fix test to fail with a helpful error message when command isn't found

* Rename and deprecate orderedCommands

* Update comment about borrowing

* Move ArtifactCommand tracking to the EngineManager trait

* Update artifact commands since tracking in the engine

* Upgrade kittycad-modeling-cmds from 0.2.85 to 0.2.86

* Remove unneeded JsonSchema derive to speed up build

* Fix to not fail on floating point differences in CI

* Update artifact commands output since truncating floating point numbers

* Fix to ensure artifact commands get cleared after a clear scene

* Update artifact commands snapshot after clearing them on clear scene

* Remove all remnants of OrderedCommands

* Update output for new simulation tests

---------

Co-authored-by: Kurt Hutten <k.hutten@protonmail.ch>

src/lang/std/engineConnection.ts

@@ -1,10 +1,10 @@
 import {
+  ArtifactCommand,
   defaultRustSourceRange,
-  defaultSourceRange,
+  ExecState,
   Program,
   RustSourceRange,
   SourceRange,
-  sourceRangeFromRust,
 } from 'lang/wasm'
 import { VITE_KC_API_WS_MODELING_URL, VITE_KC_DEV_TOKEN } from 'env'
 import { Models } from '@kittycad/lib'
@@ -20,7 +20,6 @@ import { DefaultPlanes } from 'wasm-lib/kcl/bindings/DefaultPlanes'
 import {
   ArtifactGraph,
   EngineCommand,
-  OrderedCommand,
   ResponseMap,
   createArtifactGraph,
 } from 'lang/std/artifactGraph'
@@ -37,6 +36,7 @@ import { KclManager } from 'lang/KclSingleton'
 import { reportRejection } from 'lib/trap'
 import { markOnce } from 'lib/performance'
 import { MachineManager } from 'components/MachineManagerProvider'
+import { Node } from 'wasm-lib/kcl/bindings/Node'
 
 // TODO(paultag): This ought to be tweakable.
 const pingIntervalMs = 5_000
@@ -1303,7 +1303,7 @@ export enum EngineCommandManagerEvents {
  *
  * As commands are send their state is tracked in {@link pendingCommands} and clear as soon as we receive a response.
  *
- * Also all commands that are sent are kept track of in {@link orderedCommands} and their responses are kept in {@link responseMap}
+ * Also all commands that are sent are kept track of in WASM artifactCommands and their responses are kept in {@link responseMap}
  * Both of these data structures are used to process the {@link artifactGraph}.
  */
 
@@ -1329,12 +1329,7 @@ export class EngineCommandManager extends EventTarget {
     [commandId: string]: PendingMessage
   } = {}
   /**
-   * The orderedCommands array of all the the commands sent to the engine, un-folded from batches, and made into one long
-   * list of the individual commands, this is used to process all the commands into the artifactGraph
-   */
-  orderedCommands: Array<OrderedCommand> = []
-  /**
-   * A map of the responses to the {@link orderedCommands}, when processing the commands into the artifactGraph, this response map allow
+   * A map of the responses to the WASM artifactCommands, when processing the commands into the artifactGraph, this response map allow
    * us to look up the response by command id
    */
   responseMap: ResponseMap = {}
@@ -1830,7 +1825,6 @@ export class EngineCommandManager extends EventTarget {
     }
   }
   async startNewSession() {
-    this.orderedCommands = []
     this.responseMap = {}
     await this.initPlanes()
   }
@@ -2073,28 +2067,6 @@ export class EngineCommandManager extends EventTarget {
       isSceneCommand,
     }
 
-    if (message.command.type === 'modeling_cmd_req') {
-      this.orderedCommands.push({
-        command: message.command,
-        range: sourceRangeFromRust(message.range),
-      })
-    } else if (message.command.type === 'modeling_cmd_batch_req') {
-      message.command.requests.forEach((req) => {
-        const cmdId = req.cmd_id || ''
-        const range = cmdId
-          ? sourceRangeFromRust(message.idToRangeMap[cmdId])
-          : defaultSourceRange()
-        const cmd: EngineCommand = {
-          type: 'modeling_cmd_req',
-          cmd_id: req.cmd_id,
-          cmd: req.cmd,
-        }
-        this.orderedCommands.push({
-          command: cmd,
-          range,
-        })
-      })
-    }
     this.engineConnection?.send(message.command)
     return promise
   }
@@ -2115,11 +2087,16 @@ export class EngineCommandManager extends EventTarget {
       Object.values(this.pendingCommands).map((a) => a.promise)
     )
   }
-  updateArtifactGraph(ast: Program) {
+  updateArtifactGraph(
+    ast: Node<Program>,
+    artifactCommands: ArtifactCommand[],
+    execStateArtifacts: ExecState['artifacts']
+  ) {
     this.artifactGraph = createArtifactGraph({
-      orderedCommands: this.orderedCommands,
+      artifactCommands,
       responseMap: this.responseMap,
       ast,
+      execStateArtifacts,
     })
     // TODO check if these still need to be deferred once e2e tests are working again.
     if (this.artifactGraph.size) {