Mixing Light Colors
I go with 3000K or 3500K for all around use for plant growing, but experiment with various 1750K to 6500K COBs as well (1750K is about what candle light is). In my opinion, mixing warm white and cool white LEDs in a grow light makes no sense, and I consider it a marketing gimmick at best. An exception is if you want a variable color temperature grow light, then it makes sense to to mix warm white and cool white dimmable separately, or use dimmable warm white and blue LEDs to control the color temperature.
I consider mixing red LEDs (like 630nm and 660nm, or 450nm and 470nm) to also be a marketing gimmick, unless a clear demonstration as to their combined efficacy can be demonstrated in controlled grows (temperature, humidity, CO2, and lighting levels consistent and does not significantly fluctuate to remove as many variables as possible). My first non-controlled experiments were in 2008 where I found no significant difference in 450-660, 450-630, 450-630-660nm, and white light for a leafy lettuce cultivar. I soldered up a few thousand low power 5mm LEDs to do these early experiments.
Planckian locus in the CIE 1931 chromaticity diagram (source)
There is nothing special about 6500K light for plants that may be used in vegging and don't normally use it. Higher color temperature light usually have a higher luminous efficacy, and 6500K is about the highest color temperature that is tolerated for the consumer before appearing too blue. It's more often found in work spaces. 6500K is also the color temperature of the standard illuminate D65 used in photometry. 6500K has very little to do with professional grow lighting, and traditional (non-ceramic) metal halide is 4200K.
There is nothing special about 2700K light for plants that may be used for flowering. It's about what incandescent bulbs roughly are and is close to the color temperature for the illuminate standard A used in photometry. You typically want to use this color temperature range or a bit higher for living spaces. 2700K has about 10% blue light, 4200K has about 20% blue light, 6500K has about 30% blue light. The greater the blue light content, the more compact the plant will be by reducing acid growth due to lower auxin levels.
Although we tend to use higher color temperature white light for vegging and a lower color temperature for flowering, I've gotten great veg growth with 2100K HPS for cannabis when LST (low stress training) techniques and higher lighting levels were used (500 umol/m2/sec). I've found greater growth at higher lighting levels but at lower color temperatures with various microgreens testing 2000K, 3000K, and 5000K light. If longer stems is what want (and what you get with lower lighting levels), but still want aggressive growth with larger leaves, play around with 2000K white LEDs at higher lighting levels for microgreens.