[java-inspections] PatternVariableCanBeUsed: equivalence check should be performed in narrowing cast as well

Fixes IDEA-348779 "Pattern variable can be used" inspection false positive

(cherry picked from commit a1ac8fe39011db2127d8d6d627cfae913320995a)

IJ-CR-128865

GitOrigin-RevId: 09a1a77aa76c515a5eed4af25ad03df8c2a35e86

(cherry picked from commit 42329e874ba4923e5676662bd4ffc0c7bc262d81)

IJ-CR-128986

GitOrigin-RevId: 07bc71098e0710e549289c80ad1c4392f2857259

(cherry picked from commit a4c7aaf59eb8839dbd11d5d3de528856867d4dec)

IJ-CR-128986

GitOrigin-RevId: 0e0de5be80dc8adfed8b9ed4fe5f464f2ccd191c

[inline-completion] IDEA-345516: mark `InlineCompletionGrayTextElement.Presentable.getText` as unstable

(cherry picked from commit 3d7798aab17dc65662484a2ac0487557b0afcd06)

IJ-CR-128986

GitOrigin-RevId: dd0cb472eff29ea4c14a7378f2dcc91398563fdc

(cherry picked from commit b7ae1bacab211d1e1773ce705d7026b425693b24)

IJ-CR-128716

GitOrigin-RevId: f1534c8f72a22022aa9e7b2784dd15e81657ba0c

* Instead of an allowlist in RUST_LOG env var, the blocklist is now used to mute noisy crates. It enables debugging and trace logs of FSD crates.
* All log levels inside `bootstrapOverShellSession` get the debugging level. The logs on the debug level were written but were useless. Besides, it's a too critical function for the trace level.

(cherry picked from commit 9ceb3068c59946999f3a91ccc0008e9d9e96dab2)

IJ-CR-128716

GitOrigin-RevId: dc419bfafd1a57250631beaee2f39557e3ffe04f

The previous approach with `cat` locally and `head -c $binarySize > $binaryPath` remotely turned out to be unreliable. The `head` call may have frozen infinitely. By the moment of writing this message, it was not clear why it had been failing.

This commit allows uploading an IJent binary in a different way. The interface is tied to the current implementation and aims to fixing particular bugs with uploading over WSL. A better API is to be provided later.

(cherry picked from commit dd5cf2a54193ea9054dbadc6d095ecff8b045ff8)

IJ-CR-128716

GitOrigin-RevId: 1aa6d5e2b37a75fd2eadda537e1fc1852fe25104

[java-dfa] IDEA-349093 Cast to unresolved reference inside try-catch causes 'Unreachable code' warning

(cherry picked from commit c9098e741070cddd1c89936045481521d5823bf7)

IJ-CR-128863

GitOrigin-RevId: bdb6d1beb7a015213b670f9c4bd5cd9c3e20a5e1

^KTIJ-29011 fixed

(cherry picked from commit 7de54f94b2b4e37e18e8729bfea8bcb5e6f182a1)

IJ-CR-128455

GitOrigin-RevId: 53d78ff32cd66fc0748d92c14a24ab282352576b

(cherry picked from commit abae1d7746fb900603d81372875ba2fd5c8cbeac)

IJ-MR-128940

GitOrigin-RevId: 7b8b53eb0b9cded2c8664d9cc200eaea31bab011

(cherry picked from commit 76019f7b08abb2d832efd4726d177643e73d7688)

IJ-MR-127896

GitOrigin-RevId: 91cf5f80a8c6012f0d73d9727f7872f58a5d847c

(cherry picked from commit 59c1472488546b1063b383ddae09ab2091f406cd)

IJ-MR-127896

GitOrigin-RevId: 1e3df133777834676fdc51aa61b3cd6170cc50b0

^KTIJ-28944 fixed
^KTIJ-29081 fixed
Fix for KTIJ-28801 is cherry-picked to 211

(cherry picked from commit 2cd7dae0fd76765c87d64d68b39782932e9d6cbb)

IJ-MR-127896

GitOrigin-RevId: 4e04234b479f92d7987de3a315ba91de47dbaff5

(cherry picked from commit 7525571784c1083e24e5187b285f463f5f61c4ab)

IJ-CR-128686

GitOrigin-RevId: 13580ba047e9a9dc87101fad5b8899fc4261c232

(cherry picked from commit dd0e1512677d49b87357dfd0f679644de87bf045)

IJ-CR-128686

GitOrigin-RevId: b87bda9c13af74e85e6e5656e0860d1de839803f

(cherry picked from commit 46b7b134b19f9d3c57e052d912692f9b476a346b)

IJ-CR-128686

GitOrigin-RevId: bbd32d9b75399c18d93e776d8aef20b3db4e8c55

(cherry picked from commit 0f9d90604b823c7be875651fe01574d024faedaf)

IJ-CR-128686

GitOrigin-RevId: 4a60024ffb30b09c4c82176d027c4e698a0860bd

- fix tests

(cherry picked from commit 06abdcde09eb383fd3d398df75c4d03b7fc485b0)

IJ-CR-128822

GitOrigin-RevId: 48dd4e95f52e5f2d4018d2381065ed8c78ecab57

- change level for StringConcatenationArgumentToLogCall

(cherry picked from commit a5c3f07aa9968e8a22bb58d8a3df1d62dc6c89fe)

IJ-CR-128822

GitOrigin-RevId: 4ffc4acc5832b7932df4f7c12e9a6d632ef83121

Implementation of these methods will be altered.
Fixes IDEA-347494 Unreachable code inspection triggered by quarkus base class

(cherry picked from commit 9b4f076187ad392f31a2bb1a8de5e6046defa609)

IJ-CR-128861

GitOrigin-RevId: 7e8c33db19f876ce5881a297fd70fa1a09b71b52

(cherry picked from commit d6c93af8c1ae4a421189d9bee57a7fca75116e28)

IJ-CR-127730

GitOrigin-RevId: 26a8c5a40e35de0e06922f983d7bbd57c2ce8ecb

(cherry picked from commit b57639da6cde85a1f966d533fafcfdbcbaade94f)

IJ-CR-127730

GitOrigin-RevId: 7c997c697979685943fd860173110d4a53b87cd3

Fix wording

(cherry picked from commit 30a2f640456526ad16c262b834497617ef9f1530)

IJ-CR-127730

GitOrigin-RevId: d758a9aae3ac67c94a4921bc219c8c45cc129f3f

Minor fix

(cherry picked from commit 21395d593317bd692deeb3953cdcb6ba1122b75f)

IJ-CR-127730

GitOrigin-RevId: 5b3c7ac6ac0c2c1ed00a560769cfb559e4750981

(cherry picked from commit a4c9cf9f6b86a91938b5f6a186fb8832e57e151d)

IJ-CR-127730

GitOrigin-RevId: 823ab3443db15eb42367445fbac0686923a1cd01

There was a logical bug inside `readLineWithoutBuffering`: it's impossible to distinguish an EOF looking only at `InputStream.available()`.

This problem led to awaiting a closed stream, while the error should have been returned immediately.

Also, and even more importantly, Javadoc of `available()` clarifies that there's no guarantee that the method actually returns a number of bytes available to read. That *could have been* the reason why `readLineWithoutBuffering` used to suddenly freeze.

(cherry picked from commit 857a6e9a3799093feac412329255880b57df182d)
(cherry picked from commit eebc23118666541c5dbe53e9835d561a846aa0c1)

GitOrigin-RevId: 3dc3e0081447743f3face80f9322549058ec9233

The bug existed because a child shell doesn't set the environment variable $SHELL.

```
$ echo $SHELL
/bin/bash
$ /bin/sh -c 'echo $SHELL'
/bin/bash
```

This behaviour is described in https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html :

> SHELL
> This variable shall represent a pathname of the user's preferred command language interpreter.

I.e. it should be assumed that `$SHELL == $(getent passwd "$(whoami)" | cut -d: -f7)`

(cherry picked from commit dbf778a32e8198e542792e91d40e112ded2590c4)
(cherry picked from commit 529f8d5e2ac085ee902cc8c7dd5db9e1a3732515)

GitOrigin-RevId: 00d660e6df5b67d80523f30f0ecf1b836f691942

(cherry picked from commit e265d1c6420116f000c8fb1c6f41856f8bfafbfe)

IJ-MR-128043

(cherry picked from commit ca57a13a346f3977881d598882f400c0380f61dc)

GitOrigin-RevId: 863a94c2d8a049d4c8a4344193ea90be7b6e662c

(cherry picked from commit 16f5a49227de384193be807d3b9459e4b2babc11)

IJ-CR-128752

GitOrigin-RevId: 276cd3767a7b7b89bf0388daa1110782ec4ab608

(cherry picked from commit 3dcf428ac631435614dfeabc8c3e80e7ec4ce29c)

IJ-CR-128650

GitOrigin-RevId: edc47e3408dcd6679da24fc9d3b6d96e86e688ee

ConnectException exists in java.net and in Gradle API. The method originally caught the Gradle's one, worked for Gradle. When abstracted to support Maven, it was switched to java.net.ConnectException, that worked for Maven, but failed for Gradle.
Inspecting the root cause will make it build tool agnostic.

(cherry picked from commit efc07dc7a69f258ace7cf1339a39fe1fccd36bf0)

IJ-CR-128688

GitOrigin-RevId: 7bad4907aec6e42ebe4d1742b30542eccc2abf82

IJ-CR-126919

(cherry picked from commit 84c41de2aef102a125a3fe7df12d88566292a97b)

IJ-CR-126919

GitOrigin-RevId: 6090c3d2f504bdb2aa9de4d8552c836fd27906a5

(cherry picked from commit 85690e8eb9007e4e24ed6c2a37434e12820802cf)

IJ-CR-126919

GitOrigin-RevId: bab2cfde566874fed8d9a09072f7580208abb9e0

(cherry picked from commit f5285f2c3421ab7b568ee614c6672af8b7a8c19f)

IJ-CR-126919

GitOrigin-RevId: 71048cdc14ccb9e265d313d726f8b852064eb69b

(cherry picked from commit 8f53ca0c93d576109b6f47b9dcff585b5babb91b)

IJ-CR-126919

GitOrigin-RevId: 96b950f136943c3369459e687a747490f4cc1899

GitOrigin-RevId: 23469f8019109055d6ca452769c7023e1986f883

This is an anti-pattern and can cause infinite repaint loops
(like this issue) and other problems.

The perfect fix would be to just update the preferred size
once when the state changes and cache it somewhere,
but unfortunately the size depends on the state outside
the component, for example, the number of tabs in
the tool window. It's impossible to reliably track updates
to the whole state from within the component.

So instead of a proper fix, let's work around the issue
by not calling setBorder, but instead setting up a border
with mutable insets once and then just updating those
insets. This way, we at least don't have any hidden repaint()
calls in getPreferredSize().

(cherry picked from commit fd31eba44f2c90de503f3b22210d007af09c7319)

IJ-CR-128482

GitOrigin-RevId: 5b38a62b6c3bf18f366ee97bb0c77e7ee00c0221

(cherry picked from commit 04e1d79de00128faddcefd5e5fd1c9bdc66daba5)

IJ-CR-128237

GitOrigin-RevId: 24947a453d408a38ccc4d457eb49b3f08cdb239c

(cherry picked from commit 9f2a8972a83a718c8a6c29dd89eefb4e613eaf91)

IJ-CR-127217

GitOrigin-RevId: 0987eccc214462684ef17c833a35dfadd104595a

(cherry picked from commit 3b1facb445cd5722cea7f5b65ae3d0d448cc3253)

IJ-CR-127521

GitOrigin-RevId: 36074561856ab39c77a085fe4a37162a757f5e18