The reporting list-nodes api endpoint is throwing GRPC errors well before reaching the 10k nodes-per-page limit

Describe the bug

https://docs.chef.io/automate/api/#operation/ReportingService_ListNodes says that you should be able to return up to 10,000 nodes per page in the results of your api query. However, we're seeing the following error when trying to get results for more than 5-7,000 nodes (depending on environment):

> /usr/bin/curl -s -H "api-token: XXXXXXXXX" https://automate.example.com/api/v0/compliance/reporting/nodes/search -H "Content-Type: application/json" -X POST --data '{"filters": [{"type": "","values": [""]}],"id": "","order": "ASC","page": 1,"per_page": 8000,"sort": "","type": ""}'

{"error":"grpc: received message larger than max (4471820 vs. 4194304)","code":8,"message":"grpc: received message larger than max (4471820 vs. 4194304)","details":[]}

To Reproduce

Steps to reproduce the behavior:

stand up an automate instance
use chef-load to load in 10k compliance reports
attempt to run /usr/bin/curl -s -H "api-token: XXXXXXXXX" https://automate.example.com/api/v0/compliance/reporting/nodes/search -H "Content-Type: application/json" -X POST --data '{"filters": [{"type": "","values": [""]}],"id": "","order": "ASC","page": 1,"per_page": 8000,"sort": "","type": ""}' against the automate instance

It's important to note that you need to actually have the number of compliance scan results you're trying to pull. If you set the page size to 10000 nodes-per-page, and you only have 1 node, this error won't come up. You need to actually have the 10k node compliance scan results on your automate server to replicate this error (hence using chef-load).

Expected behavior

This should return a JSON list of node objects with their compliance reporting status. It would look something like this (though much more than one node being reported):

> /usr/bin/curl -s -H "api-token: XXXXXXXXXXXXXXXXXXX" https://automate.example.com/api/v0/compliance/reporting/nodes/search -H "Content-Type: application/json" -X POST --data '{"filters": [{"type": "","values": [""]}],"id": "","order": "ASC","page": 1,"per_page": 10000,"sort": "","type": ""}'

{"nodes":[{"id":"d4dba4ee-feff-4f43-8838-ef922da33e64","name":"automate-server","platform":{"name":"centos","release":"7.7.1908","full":"centos 7.7.1908"},"environment":"automate_server","latest_report":{"id":"9666917c-3d48-4795-9167-5f138d588b91","end_time":"2021-05-04T19:06:48Z","status":"failed","controls":{"total":217,"passed":{"total":141},"skipped":{"total":1},"failed":{"total":73,"minor":1,"major":0,"critical":72},"waived":{"total":2}}},"tags":[],"profiles":[{"name":"cis-centos7-level1","version":"1.1.0-5","id":"383a52cfd11f1cf1e9ee2e54ad7b05a810e20e903d550d8f9e15550b5cb6464b","status":"failed","full":"CIS CentOS Linux 7 Benchmark Level 1, v1.1.0-5"},{"name":"linux-baseline","version":"2.2.2","id":"477f53f8f6867a0f1abe7c199a569d00c9d38d8f2a8b85dbb9cc361ca435a2b6","status":"failed","full":"DevSec Linux Security Baseline, v2.2.2"}]}],"total":1,"total_passed":0,"total_failed":1,"total_skipped":0,"total_waived":0}

If we are currently limiting the output of API queries in the API to 4MB, it would be nice if this was a configurable setting so that users could adjust that based on their needs in querying the API.

Versions (please complete the following information):

OS: MacOS and Linux making a curl call to the API
API endpoint: https://automate.example.com/api/v0/compliance/reporting/nodes/search
Automate Build Number: multiple but the latest tested on was dev branch build number 20210504113343

Aha! Link: https://chef.aha.io/epics/SH-E-508

chef / automate