The time taken (before there are diminishing returns) of extracting the herb is totally dependent on the solvent used, or in your case, the oil. If you were to use water, the answer might be days. If you used heptane, a couple of hours.
The decarboxylation and the extraction - a catch all chemical term for removal of components or a component from a particular solid liquid or gas phase, as opposed to infusion which seems to be specific to botanical extractions (so both are correct in this instance) - should be decoupled. If you start with the cannabinoidal acid and heat in solvent, you'll start extracting components at different rates, while also decarboxylating and extracting the cannabinoid. This reduces the selectivity of the extraction.
Again, depending on your chosen oil, the rate at which your desired components and undesired ones are pulled into your oil will differ. You seem to have a handle on which of your two oils is better. I would try two experiments:
Heat the dry herb to decarboxylate; less than two hours should be ample and may even be too much. Then extract/infuse your oil. Bear in mind, the more oil you use also leaves you open to extracting more rubbish you don't want, as does extending the time. You're better cutting your time down first and monitoring potency (as extracting more undesired components will lower your potency in the final extract, so potency should increase as you bring your time down) and then experimenting on reducing the oil content until your potency decreases. A decarboxylation should take about one hour at around 110C, but it depends on the amount of herb you have and if it's whole leaf, shredded, milled etc. I'm talking fine material at kilo scale. Extraction under ideal conditions is done in six hours, depending on solvent.
The second option is to extract at cooler temperatures (cool temp to 50-60C absolute maximum) first to get the cannabinoidal acids. If you have the facility, add base (diluted sodium hydroxide is ideal and would take minutes . Sodium carbonate dissolved in water may work but would be slow (days?) and you'd have to monitor carbon dioxide evolution). The acids you want will form salts and extract into the water. The neutral stuff you don't want (terpenes, waxes - long chain alkanes) stay in the oil. Separate the oil and water. You have a choice here.
- Add acid to your water layer - this may cause your cannabinoid acid to precipitate, so filter it off.
- Otherwise add fresh oil to the water, then acid and your cannabinoid acid will now be in the oil after mixing.
Whatever you choose, heat your cannabinoid acid to over 100 degrees Celsius and you should be done in an hour or two. Over 120 degrees Celsius may be too much. As I mention, some experimenting will be needed but try to ensure your herbs are consistent and not mixtures of different varieties each time.
As for your oils. One is more like a liquid "fat" - the coconut oil. The grapeseed oil is still a mixture, but is primarily "polyunsaturated" ( it's actually di-unsaturated) and about 50% bigger than the main coconut oil component. These oil mixtures are not usually used in chemical extractions as they're not pure. I assume you use digestible, non-toxic solvent oils. Otherwise I'd suggest limonene.