Integration of `fsspec`

[lobis]

This issue is to track the progress of the fsspec integration as it's going to be consisting of multiple PRs.

This work is part of an IRIS-HEP Fellowship. Here are some of the PRs performed during this period:

• feat: fsspec source implementation #692 (extensive discussion)
• feat: adding a very basic FSSpecSource #967
• feat: Use a single handler argument on uproot.reading.open #971
• feat: refactor url - object split (motivated by fsspec integration) #976
• feat: fsspec source non-blocking chunks #979
• feat: improve uproot.futures compatibility with concurrent.futures #983
• feat: remove *_handler option, fsspec source as default handler #984 (TODO: remember to merge this in the future when the deprecation warning expires).
• feat: asyncio LoopExecutor and async fsspec source #992
• chore: add types to most of the uproot.source module #996
• remove default value for hostid parameter fsspec-xrootd#26
• feat: use only loop executor for fsspec source #999
• feat: update the executor submit interface to take keyword arguments and be compatible with concurrent.futures.ThreadPoolExecutor #1001
• chore: add pre-commit formatters for toml and yaml #1002
• fix: url and object splitting for local files #1007
• feat: improve uri scheme parsing with list of available schemes from fsspec #1009
• test: local http server for tests #1010
• test: s3fs pytest unraisable exception #1012
• test: use paramiko for ssh instead of sshfs #1014
• test: cover more fsspec backends #1015
• feat: basic fsspec writing #1016
• feat: modify how multipart bytes header is built (no space) on http source #1018
• chore: replace some old code (python 2) #1020
• Fix inconsistent file mode formatting and comparison (rb+ changed to r+b) fsspec/filesystem_spec#1426
• feat: add fsspec as required dependency #1021
• feat: fsspec as required dependency #1022
• feat: set fsspec as default source #1023
• fix: s3 source options and repr #1024
• test: Emscripten build and test on GitHub Actions #1026
• test: review skipped tests (networking timeouts) #1027