Hi there
Sorry for the delayed reply, but yeah, that’s quite a few questions 
I’ll take the time to try to answer your questions today ( I’ll edit my post ) 
1 - By definition the term Mixdown means rendering multi-tracks into one, while “Mixing” is about making all tracks and elements in a song sitting well together and sound as good as it can.
You have to think about how it was done in recording studios using tape machines, basically the engineers would end up with multi-tracks recording and at some point they would combine all tracks into 1 audio file during a “Mixdown” session.
The resulting audio file would then be printed on a stereo tape ( Left Channel / Right Channel ) and sent out for “Mastering”. The first goal was to end up with a much thinner tape that could be listened on many different machines ( even consumers ones ) and much more convenient to work with for Mastering.
“Mixing” is about making all tracks and elements in a song work & sit well together. So basically it’s about levels balance, frequencies masking and phase issues, as well as “positioning” the different elements ( more front " in the face", more far and distant, panning L/R…etc).
It’s quite an exhaustive area and topic, since there’s different ways to approach mixing, some people like to mix along while they are developing the song, others prefer to do it later in the process, some people will only use tracks for each element and mix them together, others will uses groups and busses to combine drums, bass, keys, vocals…etc and of course you can use both methods at the same time.
There’s no rules to be honest, except this 1 rule : if it sounds good then it sounds good !
Aside from learning, nobody care how you get that great sounding track in the end.
You will see different approach and techniques while you watch tutorials, there’s always something to grab and learn from one producer method to another, after this it’s more workflow preferences and also what the song requires ( you won’t approach mixing a 16 tracks song the same way as a huge 100 + tracks ).
You got the idea right about the different phases of the process 
Now keep in mind that the 1st and huge headache & hassle time saver is the choice of sounds and sound design !! → It’s always better to get a recording, sample or preset that works and fit the song than trying to fix it afterwards !! So sound choice and sound design is crucial.
2 - Yes, you can use Headphones to mix and even master your tracks, but you need to have reliable headphones ( nothing is really flat except in anechoic chambers maybe ) so not adding too much bass or trebles or mid-range frequencies and also not lacking too much in those, otherwise you will try to compensate but it’s gonna sound very different on speakers, phones, car…etc.
You can use software to help with numbers like loudness and frequencies reading, but you have to know your headphones really well and keep in mind the above : in the end, what you’ve compensated is coming back into your headphones, so this can easily become a tricky loop.
Loudness, frequencies visualizers and metering software can help tho, look for ADTPR Metric A/B, Flux Analyzer or Nugen Visualizer or Mastercheck or also Excite audio Vision 4x.
There’s tutorials for some of those on SA, use the magnifier icon to search by keywords on the All Courses Page 
A good advice is to listen to your “mixdown” on different supports → Monitors/Hi-fi speakers, phone, car, Bluetooth speakers…
Some models from Sennheiser, Sony and Beyer Dynamics and Focal are good choices, a lot more expensive are headphones using Planar Magnetic drivers like Audeze and other brands, but definitely not for all budgets although they are praised by some mastering engineers.
Software like Sonarworks can help with Headphones, another approach is rooms simulation like Slate Audio VSX and also Waves plugins and other brands. Watch and read reviews, or even better, go to a music shop and bring some music you know very well and ask if it’s possible to test different Headphones models.
If you’re mixing for Dolby Atmos or multi-channels output formats like 5.1 and above, it’s more difficult to mix with headphones, however there’s also software helping with this, but again, you can totally complete a stereo mix on headphones as long as you can trust those and know them well, and it is the best approach if you can’t treat your room anyway.
3 - Yes, if you really care about your music sounding good on different listening environment, you have to create multiple mixes. It’s different to mix for a club PA system than for cutting vinyls or for digital streaming on phones and mobile devices.
Digital streaming platforms have also different loudness requirements.
There’s tutorials about this on SA and software like ADPTR Metric A/B or Nugen Audio Mastercheck can help with this ( among others ).
That said, you can get 1 average Mixdown that will translate quite well on all systems ( except for vinyls cutting which is more specific ( i.e : too loud and the needle might cut through the vinyl support. So your tracks can’t be pressed ).
One common trick is to filter out unwanted & extra low end and add saturation and harmonic distortion, but again, it’s something you’ll learn over time watching different courses and learning how different producers deal with mixing, it takes time, trial and errors and listening tests on different supports.
4 - Again, it’s about mixing on the go while you’re developing the song and adding elements and keeping balanced levels, then after this, yes, you might want to automate things depending of your arrangement.
There’s different approaches to this once again, that said, volume automation on the track fader itself is not always the best choice ( due to the resolution of DAWs faders ) and adding a utility plugin in your effect channel to control and automate volume is a common technique for this.
5 - Ultimately you only need to bounce MIDI to audio if your DAW and computer starts to suffer from high CPU loads and resources. That said, bouncing to audio can also be very handy for getting creative or if you collaborate with others that do not have the same plugins.
Working with MIDI has the advantage to be non destructive and more flexible to change sounds, presets… Bouncing to audio is more limited in this aspect, so you might want to “print” ( again think about the studios and tape recorders era , so “Bounce” audio in your DAW for both the WET and DRY track ( with or without FX ) to keep a safety net because of the destructive nature of bouncing to audio.
When using your effects as sends rather than having them on the track, you can also just send the effects returns to audio and bounce this to audio, then you’ll have the benefit to be able to mix both dry signal and wet in parallels. Some DAWs will also render all tracks including FX sends to separate audio stems. Again, the thing to keep in mind is the destructive process with bouncing to audio. Don’t hesitate to duplicate tracks and also keep different projects versions.
6 - Within the Digital domain and DAWs, output levels are in Decibels relative to full scale (dBFS or dB FS), the maximum possible level being 0 dBFS, Vu Meters on the opposite are made for the Analog domain and hardware. The 0 level for those Vu Meters correspond to a - 18 or -21 dBFS, it can help with gain staging and sending signals to plugins emulating hardware units counterparts since they are not expecting an input exceeding -18 dBFS.
That said, you might want to exceed this to drive the unit and get ( good ) saturation and distortion, and you don’t really clip a signal in the digital domain anyway, but it can help with gain staging and also appreciation of loudness. Digital meters reacts immediately to peaks, Vu meters are slower and give you a more average perceived loudness indication over time.
If you decide to base your mix on Vu meter metering, you’ll have to keep with this for all tracks, and when it comes to the final mixing stage, you can adjust this to the digital metering.
Some people only use this methods, others don’t bother at all with it, it’s something you’ll have to experience yourself and compare results.
There’s also courses about this one on SA website
Understanding Loudness and Metering with Kirk Degiorgio
Does Gain Staging Matter? with Kirk Degiorgio
Also take a look at those :
Understanding Mixing Fundamentals with Phil Johnston
Understanding Mixing Level 2 with Protoculture
7 - When sending a demo to a label, one goal is of course to get the best quality as possible. So if you’re room is a problem, you will need to compensate this, especially if there’s serious issue with the room acoustic. Headphones, listening in different environments and on different supports might help with this, of course it will be more time consuming but might worth it : labels don’t have the time and if the mix sounds bad they might not even bother to carry on after a few seconds listening.
Now there’s also another thing to consider and ask your self ( and IMHO ) this is as important and even more as a good “sounding” track ( mixing wise ) : do you think the tracks grabs attention, is it appealing for listeners, does the genre fits the label you’re sending the track to ?
If you can get other people to listen to your music and be honest about this ( is it grabbing them or not ), you don’t need an audio engineer friend for this, it’s the opposite, just get someone else listening to your music, step backward, just press play and watch their reaction : tapping feet, bouncing head and even humming the melody… Then you know you have something with this track !!
Polishing and getting it ready for delivery and submission is the technical part, and as good as you can be or get at this, the 1st essence of music is about being appealing to listeners and creativity.
A label might consider an average sounding track if the song really grabs them rather than a perfectly polished mix that’s not, but again, send the best quality you can.
Hope this helps !
We’re just scratching the surface here, nothing is written in stone when it comes to music production because it’s an art form at first, not only math and numbers.
Keep the fun, keep learning, 1 thing you were sure about for 10 years long could suddenly become irrelevant when you discover a different approach, what works best evolves with tools and techniques and in the end what matters is how your track sounds and is it grabbing listeners attention or not.
Cheers