docs: fix broken `make` in `docs/`

We had some wrapped lines without blank lines between, which made the runrst
script think the list was not a list and it got confused about the
indentation. I added blank lines, and also some other minor styling for
consistency with the rest of the file.

branchmap: use mmap for faster revbranchcache loading

A typical revbranchmap usage is:
- load the entire revbranchmap file into memory
- maybe do a few lookups
- add a few bytes to it
- write the addition to disk

There's no reason to load the entire revbranchmap into memory.
We can split it into a large immutable prefix and a mutable suffix,
and then memorymap the prefix, thus saving all the useless loading.

Benchmarking on some real-world pushes suggests that out of ~100s server-side
push handling revbranchcache handling is responsible for:

* ~7s with no change
* ~1.3s with the change, without mmap
* 0.04s with the change, with mmap

hghave: add py312 and py313

While not required in the core test suite in the moment, these could be useful
in the future or for extensions. For example, Python 3.12 removed distutils and
it might make sense to differentiate based on that.

hghave: use strings instead of floats for version numbers passed to checkvers

I think it’s a really bad idea to use floats for version numbers. One problem
is that 3.10 is the same as 3.1.

py3: fully port doctest to py3

import-checker: make stdlib path detection work in virtual environments

The previous logic tried to find the directory containing BaseHTTPServer, which
didn’t work as indended because it was only present on Python 2. Instead, the
argparse module is used now.

cleanup: remove unnecessary list constructor calls around list comprehensions

branching: merge stable into default

Added signature for changeset 3fd1efb3ad12

Added tag 6.6.3 for changeset 3fd1efb3ad12

relnotes: add 6.6.3

tests: fix nondeterministic test failure in test-contrib-perf.t

It turns out (not too shockingly!) the kernel sometimes has some work to do,
perhaps at the very least context-switching, so asserting the system time
is 0.000000 doesn't work.

grep: restore usage of --include/--exclude options

The refactor in 4a73df6eb67d accidentally forgot to transform the opts
argument for walkopts into a byteskwargs. This resulted in its options
being ignored. In particular, the -X/-I pair of options was missing.

A simple fix restores its usage. Tests included, of course.

rust-changelog: don't panic on empty file lists

tests: use sha256line.py instead of /dev/random in test-censor.t (issue6858)

Sometimes the systems that run our test suite don't have enough entropy and
they cannot produce target file of the expected size using /dev/random, which
results in test failures. Switching to /dev/urandom would give us way more
available data at the cost of it being less "random", but we don't really need
to use entropy for this task at all, since we only care if the file size after
compression is big enough to not be stored inline in the revlog. So let's use
something that we already have used to generate this kind of data in other
tests.

tests: make sha256line.py available for all tests

This was previously only used in test-revlog-delta-find.t, but it will be
useful (and used) in other tests that might need to generate
poorly-compressible files.

tests: don't use "status" operand of dd in test-censor.t (issue6858)

Some implementations don't have this operand, let's just direct stderr into
/dev/null, that's pretty cross-platform.

Also specify bs=512 (the default for me), because the default might be
different on different systems. Other uses of dd in the tests do specify it, so
this is more consistent.

delta-find: pass the full deltainfo to the _DeltaSearch class

Having more information is better, so we pass it directly.

delta-find: move sparse-revlog pre-filtering in the associated class

Lets move the specialized code in the specialized class.