It's likely to be in the task specification, sent to you in the sched_reply_... file the last time you received new work. GPUGrid does the same:

<result>
    <name>e23s14_e4s5p0f177-ADRIA_PBBandit100ns-0-1-RND9533_0</name>
    <final_cpu_time>0.000000</final_cpu_time>
    <final_elapsed_time>0.000000</final_elapsed_time>
    <exit_status>0</exit_status>
    <state>2</state>
    <platform>windows_x86_64</platform>
    <version_num>211</version_num>
    <plan_class>cuda101</plan_class>
    <report_immediately/>
    ...

It's the way the project manages its workflow. GPUGrid is doing fast-moving research, and wants its science back pronto. I suspect Primegrid may be trying to keep the size of their database of 'results out in the field' as small, and hence as cheap, as possible.

It's their science. You are volunteering to help them, not the other way round. Please respect their wishes, as far as possible.

Used to have a lodger who was into mmrpg games. He used to get shot every time an upload started. I would turn off network activity if I knew he was playing. I do the same during zoom meetings if I know an upload is due to start during the meeting as sometimes it really messes with it.
But I am guessing, that isn't the reason you might want to delay with Prime Grid as the uploads are orders of magnitude smaller.

Uploads can be multiple hundreds of MBs or GBs big. Reports are mere kilobytes, maybe low megabytes if you have a lot to report at the same time. Reporting one task at a time is a glitch on the bandwidth.

If you set your work buffer to long enough, you can actually create a schedule in settings for network access.
That way your system will only access the network at specific times.
It won't upload results, but most likely also not download new WUs either.

I just leave everything to default. My network is a 'relatively slow' 20Mbits, and most projects barely use 1-10Mbits for download.
I can still easily watch a 1440p video on one tv, and a 720/1080p video stream on another TV at the same time, without buffering issues.