Throttle block consumption (#156)

# Description
This PR adds a new optional config parameter to each network, allowing
us to define some block throttling.

the setting is called `processEveryNumBlocks` : 
* Throttle block processing to only process blocks every N blocks. Set
to 1 to process every block, 2 to process every other block, etc.

Also, note that even if we skip a block, it doesn't mean we don't have
any work on it. We still need to check if there's new orders coming in.
So what is actually skipped is only the programmatic order checks and
posting orders.

Also, it persists all Prometheus metrics, so it should count as a
"consumed block" for our consumption rate alerts, meaning we should see
the block consumption rate progressing faster.

<img width="1708" alt="image"
src="https://github.com/user-attachments/assets/3e4e9b96-c141-416d-b399-e315abc355f8">



This is an example on how to use it: `"processEveryNumBlocks": 40,`
This would do polling and post orders once every 40 blocks. In arbitrum
this should be every 8 seconds

```yaml
{
  "networks": [
    {      
      "name": "arbitrum-one",
      "rpc": "wss://your-rpc/arbitrum",
      "deploymentBlock": 204704802,
      "processEveryNumBlocks": 40,
      "filterPolicy": {
        "defaultAction": "DROP",
        "handlers": {
          "0x44569Cbd4E10dd5e97293337964Eff32d58ed352": "ACCEPT",
          "0x519BA24e959E33b3B6220CA98bd353d8c2D89920": "ACCEPT",
          "0x6cF1e9cA41f7611dEf408122793c358a3d11E5a5": "ACCEPT",
          "0xd3338f21c89745e46af56aeaf553cf96ba9bc66f": "ACCEPT",
          "0xE8212F30C28B4AAB467DF3725C14d6e89C2eB967": "ACCEPT",
          "0xB148F40fff05b5CE6B22752cf8E454B556f7a851": "ACCEPT"
        }
      },
      "watchdogTimeout": 120
    }
  ]
}
```

Dockerfile

@@ -1,4 +1,4 @@
-FROM node:alpine AS build
+FROM node:22.2-alpine AS build
 
 WORKDIR /usr/src/app
 

src/services/chain.ts

@@ -77,6 +77,8 @@ export class ChainContext {
   readonly dryRun: boolean;
   readonly watchdogTimeout: number;
   readonly addresses?: string[];
+  readonly processEveryNumBlocks: number;
+
   private sync: ChainSync = ChainSync.SYNCING;
   static chains: Chains = {};
 
@@ -106,6 +108,7 @@ export class ChainContext {
     this.deploymentBlock = deploymentBlock;
     this.pageSize = pageSize ?? PAGE_SIZE_DEFAULT;
     this.dryRun = dryRun;
+    this.processEveryNumBlocks = options.processEveryNumBlocks ?? 1;
     this.watchdogTimeout = watchdogTimeout ?? WATCHDOG_TIMEOUT_DEFAULT_SECS;
     this.addresses = owners;
 
@@ -168,7 +171,7 @@ export class ChainContext {
    * @returns the run promises for what needs to be watched
    */
   public async warmUp(oneShot?: boolean) {
-    const { provider, chainId } = this;
+    const { provider, chainId, processEveryNumBlocks } = this;
     const log = getLogger("chainContext:warmUp", chainId.toString());
     let { lastProcessedBlock } = this.registry;
     const { pageSize } = this;
@@ -213,7 +216,13 @@ export class ChainContext {
           log.info(
             `🔄 Start sync with from block ${fromBlock} to ${toBlock}. Pending ${
               toBlock - fromBlock
-            } blocks (~${Math.ceil((toBlock - fromBlock) / pageSize)} pages)`
+            } blocks (~${Math.ceil(
+              (toBlock - fromBlock) / pageSize
+            )} pages, processing every ${
+              processEveryNumBlocks > 1
+                ? processEveryNumBlocks + " blocks"
+                : "block"
+            })`
           );
         }
 
@@ -311,9 +320,8 @@ export class ChainContext {
     let lastBlockReceived = lastProcessedBlock;
     provider.on("block", async (blockNumber: number) => {
       try {
-        const block = await provider.getBlock(blockNumber);
-
         log.debug(`New block ${blockNumber}`);
+        const block = await provider.getBlock(blockNumber);
 
         // Set the block time metric
         const _blockTime = block.timestamp - lastBlockReceived.timestamp;
@@ -434,7 +442,7 @@ async function processBlock(
   blockNumberOverride?: number,
   blockTimestampOverride?: number
 ) {
-  const { provider, chainId } = context;
+  const { provider, chainId, processEveryNumBlocks } = context;
   const timer = metrics.processBlockDurationSeconds
     .labels(context.chainId.toString())
     .startTimer();
@@ -463,24 +471,29 @@ async function processBlock(
     }
   }
 
-  // run action
-  const result = await checkForAndPlaceOrder(
-    context,
-    block,
-    blockNumberOverride,
-    blockTimestampOverride
-  )
-    .then(() => true)
-    .catch(() => {
-      hasErrors = true;
-      log.error(`Error running "checkForAndPlaceOrder" action`);
-      return false;
-    });
-  log.debug(
-    `Result of "checkForAndPlaceOrder" action for block ${
-      block.number
-    }: ${_formatResult(result)}`
-  );
+  // Decide if we should process this block
+  const shouldProcessBlock = block.number % processEveryNumBlocks === 0;
+
+  // Check programmatic  orders and place orders if necessary
+  if (shouldProcessBlock) {
+    const result = await checkForAndPlaceOrder(
+      context,
+      block,
+      blockNumberOverride,
+      blockTimestampOverride
+    )
+      .then(() => true)
+      .catch(() => {
+        hasErrors = true;
+        log.error(`Error running "checkForAndPlaceOrder" action`);
+        return false;
+      });
+    log.debug(
+      `Result of "checkForAndPlaceOrder" action for block ${
+        block.number
+      }: ${_formatResult(result)}`
+    );
+  }
 
   timer();
   if (hasErrors) {

src/types/config.schema.json

@@ -20,6 +20,12 @@
           "watchdogTimeout": {
             "type": "integer"
           },
+          "processEveryNumBlocks": {
+            "type": "integer",
+            "minimum": 1,
+            "description": "Throttle block processing to only process blocks every N blocks. Set to 1 to process every block (default), 2 to process every other block, etc.",
+            "default": 1
+          },
           "orderBookApi": {
             "type": "string",
             "format": "uri"

src/types/types.d.ts

@@ -13,6 +13,10 @@ export interface Config {
     rpc: string;
     deploymentBlock: number;
     watchdogTimeout?: number;
+    /**
+     * Throttle block processing to only process blocks every N blocks. Set to 1 to process every block, 2 to process every other block, etc.
+     */
+    processEveryNumBlocks?: number;
     orderBookApi?: string;
     pageSize?: number;
     filterPolicy: {