Docs: try to make model options / model install methods clearer

[mrienstra]

Primary goal: make it easy to -- from the "Quick start" section -- get info about models by following a link to ./models/README.md, and then quickly find answers to some basic questions about the various models, i.e. the meaning of .en, -q5_0, & -tdrz. And link upstream (to openai/whisper) to see their info on the various models.

Related issues: #1396, #1076

Aside: Noticed some inconsistencies along the way, I can back those changes out and put them in a different PR if they make this PR too unwieldy to review.

1. ggml was most often wrapped in backticks, but not always, fixed that.
2. Fixed some hard-wrapped lines (in .md files), as non-hard-wrapped seemed to be the convention.
3. Fixed code blocks language identifiers, used bash, text (for mixture of shell input & output), or powershell, primarily in place of java or non-specified.
4. Some formatting fixes were made by Prettier (running in VS Code), eg. properly padding tables, changing double spaces to single spaces, removing trailing spaces, changing * bullets to - (which was already most often the case), changing [X] checkmarks to [x] (this was a toss-up, it was 50/50 before, but at least now the default VS Code Markdown formatter won't try to change them?). There were a few changes Prettier made that I backed out, e.g. changing * for italics to _.
5. Fixed missing $ shell prompts, for consistency (was at 4/8, now 8/8, unless I missed some).
6. Changed a // comment to a # comment (in a Bash code block), per convention.

Actual changes of substance:

./README.md:

Before: README.md#quick-start
After: README.md#quick-start

Minor clean-up of "Quick start" instructions. Mostly wanted to make it obvious where to go to learn about the available models (./models/README.md).

./models/README.md:

Before: models/README.md
After: models/README.md

1. Opening section ended with "A third option to obtain the model files is to download them from Hugging Face:", but the "3 options" were kind of jumbled together / interwoven, so decided to spell them out one-by-one.
2. In the "Available models" section: (A) Added more models to the table. (B) Added a paragraph below the table, briefly explaining the model naming conventions, and linking to relevant documentation.

./models/download-ggml-model.sh:

Before: models/download-ggml-model.sh
After: models/download-ggml-model.sh

1. Re-ordered list of models, for consistency with the order used in ./models/README.md, e.g. tiny before tiny.en.
2. Added large-v2-q5_0 to the list.
3. Changed usage out, see below.

Bash usage output, before (screenshot is 80 characters wide, 4 lines wrapping to 7 lines):

$ bash ./models/download-ggml-model.sh              
Usage: ./models/download-ggml-model.sh <model> [models_path]

  Available models: tiny.en tiny tiny-q5_1 tiny.en-q5_1 base.en base base-q5_1 base.en-q5_1 small.en small.en-tdrz small small-q5_1 small.en-q5_1 medium medium.en medium-q5_0 medium.en-q5_0 large-v1 large-v2 large-v2-q5_0 large-v3 large-v3-q5_0

$

Bash usage output, after (screenshot is 80 characters wide, 11 lines, no wrapping):

$ bash ./models/download-ggml-model.sh
Usage: ./models/download-ggml-model.sh <model> [models_path]

Available models:
  tiny tiny.en tiny-q5_1 tiny.en-q5_1
  base base.en base-q5_1 base.en-q5_1
  small small.en small.en-tdrz small-q5_1 small.en-q5_1
  medium medium.en medium-q5_0 medium.en-q5_0
  large-v1 large-v2 large-v2-q5_0 large-v3 large-v3-q5_0

___________________________________________________________
.en = english-only -q5_[01] = quantized -tdrz = tinydiarize
$

Remaining questions

Edit: moved my remaining questions to a new issue: #1807