Remove context-specific parse

[drhagen]

Inspired by #77, I found that removing the parse_method global mutable state is relatively straightforward. It can be replaced by a switch that chooses the Reader subtype based on the input type. All the readers seem to work in all contexts, so this is a change that dramatically reduces the complexity of the context system.

This changes the error messages when parsing str input in a GeneralParsers context, but that seems like a good thing overall.

[codecov-commenter]

Codecov Report

Merging #78 (662f0ed) into master (dba1b6e) will not change coverage.
The diff coverage is 100.00%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the Github App Integration for your organization. Read more.

@@            Coverage Diff            @@
##            master       #78   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           25        25           
  Lines          836       826   -10     
  Branches       100       102    +2     
=========================================
- Hits           836       826   -10     

Impacted Files	Coverage Δ
src/parsita/metaclasses.py	100.00% <ø> (ø)
src/parsita/options.py	100.00% <100.00%> (ø)
src/parsita/parsers/__init__.py	100.00% <100.00%> (ø)
src/parsita/parsers/_base.py	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[drhagen]

@lostinplace My idea to remove the parse_method global state seems to have worked. But I don't want to merge this without you taking a look at it. I'll want to merge this and then review your BytesReader PR on top of it because this should simplify it quite a bit.