chore: regenerate README (#67)

This PR was generated using Autosynth. 🌈


<details><summary>Log from Synthtool</summary>

```
2021-02-09 19:02:59,946 synthtool [DEBUG] > Executing /root/.cache/synthtool/java-pubsublite-spark/.github/readme/synth.py.
On branch autosynth-readme
nothing to commit, working tree clean
2021-02-09 19:03:00,838 synthtool [DEBUG] > Wrote metadata to .github/readme/synth.metadata/synth.metadata.

```
</details>

Full log will be available here:
https://source.cloud.google.com/results/invocations/5f590b49-f4bb-4b58-ad11-94fd7e8e7ae1/targets

- [ ] To automatically regenerate this PR, check this box.

.github/readme/synth.metadata/synth.metadata

@@ -4,7 +4,7 @@
       "git": {
         "name": ".",
         "remote": "https://github.com/googleapis/java-pubsublite-spark.git",
-        "sha": "f19ac3f62de6246347a9e2b64a4bb074d430a228"
+        "sha": "5d9fb4a8ffc8daa7b30b8c9c536bd49ed54f2acc"
       }
     },
     {

README.md

@@ -116,12 +116,13 @@ and manual Spark installations.
 
 The connector supports a number of options to configure the read:
 
-  | Option | Type | Required | Meaning |
-  | ------ | ---- | -------- | ------- |
-  | pubsublite.subscription | String | Y | Full subscription path that the connector will read from. |
-  | pubsublite.flowcontrol.byteoutstandingperpartition | Long | N | Max number of bytes per partition that will be cached in workers before Spark processes the messages. Default to 50000000 bytes. |
-  | pubsublite.flowcontrol.messageoutstandingperpartition | Long | N | Max number of messages per partition that will be cached in workers before Spark processes the messages. Default to Long.MAX_VALUE. |
-  | gcp.credentials.key | String | N | Service account JSON in base64. Default to [Application Default Credentials](https://cloud.google.com/docs/authentication/production#automatically). |
+  | Option | Type | Required | Default Value | Meaning |
+  | ------ | ---- | -------- | ------------- | ------- |
+  | pubsublite.subscription | String | Y | | Full subscription path that the connector will read from. |
+  | pubsublite.flowcontrol.byteoutstandingperpartition | Long | N | 50_000_000 | Max number of bytes per partition that will be cached in workers before Spark processes the messages. |
+  | pubsublite.flowcontrol.messageoutstandingperpartition | Long | N | Long.MAX | Max number of messages per partition that will be cached in workers before Spark processes the messages. |
+  | pubsublite.flowcontrol.maxmessagesperbatch | Long | N | Long.MAX | Max number of messages in micro batch. |
+  | gcp.credentials.key | String | N | [Application Default Credentials](https://cloud.google.com/docs/authentication/production#automatically) | Service account JSON in base64. |
 
 ### Data Schema